Skip to main content

Full text of "Pelagius's expositions of thirteen epistles of St. Paul .."

See other formats

i}.. ;; 




;;;':;!!, .".'I^^^HBHI^^H 

1' ::;;-;' ''-'titif^^^H 













C. F. CLAY, Manager 
















M.A. (OXON.) D.LITT. (aBEED.) 






ni\^^>A ^ 


■ / 




















A NATURAL sequel to A Study of Ambrosiaster in the present 
series (volume vil, part 4, 1905) would have been a critical 
edition of his commentary on the Epistles of St Paul ; but the task 
of preparing such an edition had already been assigned by the 
Vienna Academy of Sciences to my friend Father Brewer, S.J. It 
was therefore necessary that I should turn to something else. 
When casting about in 1904 for an unworked field of research, 
I received the same suggestion independently from two scholars, 
the Editor of this series and Dr C. H. Turner, that I should edit 
the commentary of Pelagius on the Epistles of St Paul. The 
suggestive book of the late Professor Heinrich Zimmer of Berlin, 
Pelagius in Irland, published in 1901, had provided certain 
materials, and furnished some hints as to how the original com- 
mentary of Pelagius could be reconstructed, or partially recon- 
structed, from the works of later authors who had used it. I had 
not then the slightest expectation that three copies of the com- 
mentary in its original form would fall into my hands, one in 1906 
and two in 1913: yet such was to be the case. 

The fortunate discoveries, however, did not, I felt, exempt 
me from the duty of reconstructing textually the archetypes of 
authorities like Pseudo-Jerome, originally edited by Erasmus in 
1516, and Pseudo-Primasius, first published by Gagney in 1537, 
who had incorporated the greater part of Pelagius's work in their 
own\ For it was obvious that these two authorities had employed 
other copies of the original Pelagius than those that were in my 
hands, and that by their aid I could control the texts I had found. 
Further, certain other writers, like Sedulius Scottus whose 
commentary was first published in 1528, and Zmaragdus whose 
compilation was first printed in 1536, had made considerable use of 
Pelagius, and had copied his work in general with accuracy. Their 

1 It gave me the intensest satisfaction to be able, in 1906, to restore the Pseudo- 
Primasius to Cassiodorus and his pupils. 


quotations from Pelagius had in consequence to be restored to 
their original form, as far as it was possible for me to do so. One 
could also have gone on exploring the mediaeval commentaries and 
publishing certain of them (like most of those of Claudius of 
Turin, which still await an editor), but I have left such work, if it 
be thought necessary, to my successors. They will find in these 
volumes^ I trust, a sufficiently secure basis for further enquiry. 

It is obvious that the preparation of an edition like the present, 
in which an endeavour has been made to repair the undeserved 
neglect of four centuries, has cost much money, time and trouble. 
I cannot sufficiently express my gi'atitude to the Trustees or 
Managers of the following institutions or funds for the ungrudging 
confidence and lavish help extended to the researcher from 1906 
to 1915: the Hort Fund at Cambridge; the Revision Surplus 
Fund at Oxford; the Schweich Fund of the British Academy; 
Magdalen College, Oxford ; and the Carnegie Trust for the Uni- 
versities of Scotland. The work has entailed nine journeys on the 
Continent, which occupied fourteen months in all. The collations 
were made with the utmost fullness I could attain, in order that 
I might learn the exact relationship between the manuscripts, and 
represent in my critical apparatus the readings of archetypes 
rather than those of individual codices. I was desirous also to 
write part of the extraordinary history of Pelagius's commentary 
in the Middle Ages, and to do this properly required the pre- 
paration of a critical edition of the interpolations foisted on it. 

The book could never have been completed without the 
generous help of many scholars at home and abroad. It would not 
be fitting to record here the names of leading biblical, patristic and 
palaeographical authorities of our time, to Avhom I have submitted 
my various difficulties as' they arose. I have tried to specify each 
obligation in its proper place in the body of the work. To the 
librarians and officials at the libraries of Aberdeen, Arras, Bamberg, 
Basle, Berlin, Cambridge (University, Corpus, St John's), Dublin 
(Trinity College), Einsiedeln, Florence (Laurentian), Gotha, 
Grenoble, The Hague (Royal Library, Museum Meermanno- 

1 The second volume, containing text, critical apparatus and indexes, should 
appear about a year after this. The third, containing the interpolations, is post- 
poned till the an-ival of better economic conditions. 


Westreenianum), Karlsruhe, London(British Museum), Luxemburg, 
Manchester (John Rylands), Milan, Munich (State and Univer- 
sity), Nurnberg (Stadtbibliothek, Germanisches Museum), Oxford 
(Bodleian, Balliol, Magdalen, Mansfield, Merton), Paris (Biblio- 
theque Nationale, Mazarine, Ste Genevieve), Rome (Vatican, 
Basilicana, Angelica, Casanatense, Vallicelliana, Vittorio-Em- 
manuele), St Gall (Stiftsbibliothek, Stadtbibliothek), Salisbury, 
Troyes, and Zurich (Kantonsbibliothek, Stadtbibliothek) my 
heartiest thanks are due. Nor must I forget the friends in 
various countries whose gracious hospitality cheered the exile's 

The commentary here printed for the first time in its original 
form happens to be the earliest extant work by a British author^ 
Theologians will perhaps be glad to see the earliest Pelagian docu- 
ment as it first appeared. It also contains within it, if I am right, 
a form of Old-Latin text of the Epistles of St Paul read by our 
ancestors of the British Church two centuries before Augustine 
ruled the Province of Canterbury. The second volume will shed 
some light on the history of the Vulgate of St Paul's Epistles. 

I am deeply conscious of the defects of my work. The mass of 
material collected has proved immensely difficult to control, and 
the arrangement of it for press has been attended by many vexatious 
interruptions. Yet the book contains some new things, and my 
task was well worth attempting. Best thanks are due not only 
to the Editor of the series but to all who have taken part in the 
printing of the book, for the valuable help they have rendered. 



March 6th, 1922. 

The Cambridge History of English Literature, vol. i (1908), p. 65 (M. E. .James). 



I. Pelagius and his Commentary. Introduction. A Kecord of 

Previous Research 1 

The Name 'Pelagius' 1 

Pelagiiis the Briton or the Irishman 2 

His Commentary on thirteen Epistles of St Paul .... 3 

History of Pseudo- Jerome Commentary in print .... 6 

II. How to identify the Pelagius Commentary .... 34 

The Vatican Fragments ........ 48 

Interpolation in Certain MSS of Ambrosiaster on First and Second 

Corinthians ........... 51 

The Cassiodorian Commentary (Pseudo-Primasius) ... 60 

The Extracts from John the Deacon 61 

Later Compilations 63 

III. The Whole Commentary the Work of one Author . . 64 

(a) Cross References from one part of the Commentary to another 64 

(b) Illustrations of Method of Exegesis in General ... 65 
(cj Community of Ideas throughout ...... 69 

(d) Favourite Verses of Scripture 74 

(e) Community of Style and Language 79 

1. Grammar 80 

1. Accidence 80 

2. Syntax 80 

II. Lexicography. Details of Phraseology and Vocabulary . 85 

(a) Favovu-ite Openings of Notes 85 

(6) Similar Phrases in the Body of the Notes, and Phrases 

introducing Biblical Quotations ..... 89 

(c) Characteristic Words and Plirases, alphabetically ar- 
ranged .......... 92 

Supplementary Note . .115 

The Authenticity of the Prologue and Arguments . . . 115 




The Biblical Texts used by Pelagius 


§ 1. The Text of the Pauline Epistles 

(a) Occasional references to variae lectiones by Pelagius himself 
(6) Quotations from the Epistles made in the body of the notes 
(c) The light thrown by the comments on the character of the 

text which lay before the author 
Vatican Fragments 
Interpolation in Ambrosiaster . . 
External Confirmation of the use of the D type of text by 
Pelagius. Epistula ad Demetriadem 

De I)iduratione Cordis Pharaonis 

Relation of the Pauline text used by Pelagius to the quota 

tions in Gildas (a.d. 500 — 570) 

Ambrose's disagreements with the Vulgate, paralleled by 

Pelagius's text 

Appendix to § 1. Pelagius and the Vulgiite of the Pauline Epistles 
§ 2. The Text of the Other Parts of Scriptui-e 

The Heptateuch 

The Historical Books 

Psalms .... 
Sapiential Books 
The Minor Prophets 
The Major Prophets 
The Gospels . 
Acts .... 
The Epistle to the Hebrews 
Canonical Epistles . 
Apocalypse . . . 

V. Notes on the Sources used in the Commentary 
Jerome .... 

Origen-Rufinus on Romans 
Chrysostom . . . 
Theodore of Mopsuestia 
Subsidiary Sources 





VI. The Materials for the Reconstruction of the Text of the 
Commentary, and their Interrelations . 

Introduction . . 

12 ( 

(a) The Manuscripts of the Original Form . 

(1) Codex Augiensis CXIX at Karlsruhe (A) 

(2) Codex Collegii Balliolensis Oxon. 157 (B) 

(3) Codex Collegii Mertonensis Oxon. 26 (O) . 

(4) The Vatican Fragments (.Jiv) • 

(5) The Freiburg Fragments (K) . 

(b) Manuscripts of Interpolated Forms 

(1) No. 73 in the Stiftsbibliothek at St Gall (G) 

(2) No. 653 in the Biblioth^que Nationale at Pari 
The Pseudo-Jerome Manuscripts (H) 

(3) Paris, B.N. 9525 (E) . 
{Z^) The Editio Princeps .... 

(4) Salisbury, Cathedral Library, no. 5 (S) 

(5) Munich, Staatsbibliothek, lat. 13038 (R) 

(6) Munich, Universitatsbibliothek MS in fol 

(7) Paris, B.N. 1853 (M). 

(8) Spinal, No. 6 (N) . 

(9) The lost MS used by the corrector of R 

(10) Troyes, 486 (C) 

(11) Florence, R. Bibl. Mediceo-Laur. Plut. xv dext, 

(12) Cambridge, University Library, Ff. 4. 31 
Cassiodorus (Pseudo-Primasius) : Revision of Pelagius 
Wurzhiu-g (Wb) and other Glosses 
Claudius of Turin . 
Zmaragdus of St Mihiel 
Sedulius Scottus . 
Haymo of Auxerre 


Note on Prologues or Arguments 
Summary .... 
Supplementary note 

Indexes : (1) Names and Matters 

(2) Scripture References (Text or 

(3) Latin words . 

(4) Manuscripts cited . 

(5) Modern Authorities 












A. Bruckner, Quellen zur Geschichte des Pelagianischen Streites (Tubingen, 

D. DE Bruyne, 'Le Prologue Inedit de Pelage k la Premiere Lettre aux 

Corinthiens' {Revue Benedictine xxiv [1907] pp. 257-263). 

D. DE Brutne, 'Etude sur les Origines de notre Texte Latin de Saint Paul' 

{Revue Biblique nouv. ser. xii [1915] pp. 358-392). Cf. Revue B4nedictine 

XXXIII (1921) Bull. pp. 6-9. 
J. B. Burt, 'The Origin of Pelagius' {Hermathena xiii [1905] pp. 26-35). 
J. Chapman, 'Cassiodorus and the Echtemach Gospels' {Revue Benedictine 

xxviii [1911] pp. 283-296). 
H. Denifle, Die ahendldndischen Schriftausleger bis Luther liber Justitia Dei 

(Rom. 1, 17) und Justificatio, Beitrag zur Geschichte der Bxegese, der 

Literatur und des Dogmas im Mittelalter (Mainz, 1905). 

E. V. DoBSCHiJTZ, 'Ein Biicherkleinod ' pp. 18 {Jahresbericht der Schles. Gesell- 

schaftfur vaterl. Cidtur 1913). 
M. EsPOSlTO, 'A Seventh-Century Commentary on the Catholic Epistles' 

{Journal of Theological Studies xxi [1919-20] pp. 316-318). 
J. GwYNN, Liber Ardmachanus : The Book of Armagh, edited with introduction 

and appendices (Dublin and London, 1913). 
S. Hellmann, Sedidius Scotttcs (Miinchen, 1906 [published 1905]). 
P. Lehmann, Johannes Sichardus und die von ihm benutzten Bibliotheken und 

Handschriften (Miinchen, 1911). 
P. Lehmann, ' Cassiodorstudien, vii. Der Romerbrief Kommentar ' {Fhilologus 

Lxxiv [1917] pp. 354-356). 

F. LooFS, 'Pelagius, gest. nach 418, und der pelagianische Streit' {Realen- 

cyklopddie filr protestantische Theologie und A'iVcAe...Herzog...Hauck, xv 
[Leipzig, 1904] pp. 747-774) (with an extensive bibliography). 
F. LooFS, Leitfaden zum Studium der Dogmengeschichte 4^ Aufl. (Halle, 1906). 

F. LooFS, 'Pelagius' (^eaZertC3/H....Herzog...Hauck, xxiv [Leipzig, 1913] pp. 310 

-312) (with an extensive bibliography of the recent period). 
E. Mangenot, ' Saint Jerome ou Pelage ^diteur des Epltres de Saint Paul dans 

la Vulgate' pp. 37 {Revue du Clerge fran^ais, 1916). 
G.Mercati, 'Some New Fragments of Pelagius' {J.T.S. viii [1906-7] pp. 526- 


G. MoRiN, ' Jean Diacre et le Pseudo- Jerome sur les Epltres de S. Paul ' {Revue 

Benedictine xxvii [1910] pp. 113-117). 

^ For the older literature especially, see under ' F. Loofs ' in this bibliography. 


G. MoRiN, 'Un Traitd Inedit attribu^ k Saint Augustin, le de viii Qiiaestionihus 

ex Vet. Test, du Catalogue de Lorscli' {R.B. xxviii [1911] p. 3). 
G. MoRiN, 'A Propos du Quicumque : Extraits d'Homelies de S. C^saire d' Aries 

sous le nom de S. Athanase' {KB. xxviii [1911] pp. 420-421). 
G. MoRiN, Etudes, Textes, Decoiivertes: Contributions a, la Litterature et d 

VHistoire des douze premiers siecles t. i (Maredsous and Paris, 1913), 

pp. 23, 345 especially. 
E. Riggenbach, Unbeachtet gehliebene Fragmente des Pelagius-Kommentars 

(Gutersloh, 1905). 
E. Riggenbach, Diedltesten lateinischenKommentare zum ffebraer brief (Lei])zig, 

E. Riggenbach, 'Eine wichtige Entdeckung fiir die Pelagius-forschung ' (Theo- 

logisches Literaturblatt xxviil [1907] pp. 73-75). 
E. Riggenbach, 'Neues liber Pelagius' {Theologisches Literaturblatt xxvm 

[1907] p. 425). 
H. V. Schubert, Der sogenannte Praedestinatus, ein Beitrag zur Geschichte des 

Pelagianismus (Leipzig, 1903). 
Alfred J. Smith, ' The Latin Sources of the Commentary of Pelagius on the 

Epistle of St Paul to the Romans' {J.T.S. xix [1917-18] pp. 162-230; 

XX [1918-19] pp. 55-65, 127-177). 
A. Souter, 'The Commentary of Pelagius on the Epistles of St Paul' {The 

Expositor 1907 i pp. 455-467). (Written May, 1906.) 
A. SoUTER, ' Prolegomena to the Commentary of Pelagius on the Epistles of 

St Paul {J.T.S. VII [1905-6] pp. 568-575). 
A. SouTER, 'The Commentary of Pelagius on the Epistles of Paul : the Problem 

of its Restoration ' {Proceedings of the British Academy ll [1905-6] pp. 409- 

439). (Read Dec. 12, 1906.) 
A. SouTER, ' The Relation of the Roman Fragments to the Commentary in 

the Karlsruhe MS (Augiensis cxix)' {J.T.S. viii [1906-7] pp. 535-536). 
A. Souter, 'Contributions to the Criticism of Zmaragdus's Expositio Libri 

Comitis' {J.TS. ix [1907-8] pp. 584-597). 
A. Souter, 'An unrecorded Reference to the Rides of Tyconius' {J.T.S. xi 

[1909-10] pp. 562-563). 
A. Souter, 'Another New Fragment of Pelagius' {J.T.S. xii [1910-11] 

pp. 32-35). 
A. Souter, 'An Interesting Latin Subscription' (accompanied by a note from 

J. Loth) {Revue Celtique xxxil [1911] pp. 152-153). 
A. Souter, ' Freiburg Fragments of a MS of the Pelagian Commentary on the 

Epistles of St Paul' {J.T.S. xiii [1911-12] pp. 515-519). 
A. Souter, ' Cassiodorus's Copy of Eucherius's Instructiones' {J.T.S. xiv 

[1912-13] pp. 69-72). 
A. Souter, 'Dismembered Manuscripts' {R.B. xxix [1912] pp. 367-368). 
A. Souter, ' The Commonitorium of Fulgentius of Ruspe on the Holy Spirit ' 

{J.T.S. XIV [1912-13] pp. 481-^ 


A. SouTER, *New Manuscripts of Pelagius' {Theologische Literaturzeitung, 

XXXVIII [1913] p. 442). 
A. SouTER, 'Pelagius and the Pauline Text in the Book of Armagh' {J.T.S. 

XVI [1914-15] p. 105). 
A. SoUTER, ' Pelagius' Doctrine in Relation to his early life ' ( The Expositor 

1915 I pp. 180-182). See p. 3 n. 5. 
A. SouTER, 'A Theological Tractate on the Divinity of the Son, from Paris MS 

B.N. Lat. 653' (J. r.^. xvii [1915-16] pp. 129-136). 
A. SouTER, 'The Character and History of Pelagius' Commentary on the 

Epistles of St Paul' {Proceedings of the British Academy vii [1915-16] 

pp. 261-296). (Read March 15, 1916.) 
A. SouTER, ' Pelagius' Text of Romans v 12, with Comment ' {Expository Times 

xxviil [1916-17] pp. 42-43). 
A. SouTER, ' The Sources of Sedulius Scottus' Collectaiieum on the Epistles of 

St Paul' {J.T.S. XVIII [1916-17] pp. 184-228). 
A. SoDTER, 'The Earliest Siu-viving Book of a British Author' {The Con- 
temporary Review cxv [1919 i] pp. 76-82). 
A. SoUTER, ' Further Contributions to the Criticism of Zmaragdus's Expositio 

Libri Comitis' {J.T.S. xxiii [1921-22] pp. 73-76). 
L. C. Stern, Epistolae Beati Pauli glosatae glosa interlineali (Halle, 1910). 
C. H. Turner, 'Pelagius' Commentary on the Pauline Epistles and its History' 

{J.T.S. IV [1902-3] pp. 132-141). 
A. Vaccari, Un Commenio a Giobbe di Giuliano di Eclana (Roma, 1915). 
H. Zimmer, Pelagiiis in Irland: Texte und Untersnchungen zur patristischen 

Litteratur (Berlin, 1901), 



The Name ' Pelagius.' The name Pelagius is obviously Greek 
in ultimate origin, being derived from TreXaYo?, ' the sea.' To judge, 
however, by the Greek Onomasticon and indexes to inscriptions and 
papyri, it appears to have been comparatively rare in that language \ 
The oldest instance of the name that I can find in Greek characters 
is in an inscription of Halaesa in Sicily, belonging to the period 
of the Roman republic-. The name, meaning Seaman, appears 
to have had no vogue in the Greek world until about the second 
or third century of our era. But already in the first century before 
Christ the common noun TreXayo'i, along with its derivative adjec- 
tive 7r€\dyio<;, had become a Latin word also. It doubtless became 
familiar to Westerns, at first, as a name sometimes borne by Greek 
slaves. As such, it was kept as a cognomen or an agnomen on 
manumission. And in fact there are at least three clear instances 
of it with this value in Latin inscriptions^. In other inscriptions it 
is found standing by itself^ The oldest dated Latin example of the 
name is in an inscription of a.d. 145 ^ Both in Greek and in Latin 
there are a good many instances of the feminine HeXayla (Pelagia) 
also. The Dictionary of Christian Biography knows a dozen persons 
with the name Pelagius, besides one or two with the name Pelagia. 

^ W. Pape's Worterbuch der griechischen Eigennamen 3 Aufl. bearb. v. G. E. 
Benseler (Braunschw. 1884) s.v. gives only three or four examples, none of which is 
perhaps earlier than the fourth century of our era. See also Inscriptiones Graecae 
vol. Ill p. 2 (Berol. 1882) 3439 (Attica and Christian) ; vol. xiv (Berol. 1890) 352 
(Halaesa in Sicily); Oxyrhynchus Papyri 43 (a.d. 295). 

- See note 1. 

3 C.I.L. VIII 9689 (Cartenna in Mauretania) ; ix 3941 (Alba Fucens in Italy) ; xiv 
1119 (Ostia). 

•* E.g. C.I.L. II 5812 (Segisamo in Spain: a.d. 239); x 670 (Salernum in Italy); 
2038 (Puteoli); it seems to be a 7iomen in C.I.L. xii 1815 [ = Dessau, Tnscr. Lat. Sel. 
7264] (Vienne in France) Ant. Pelagius. 

s C.I.L. Ill 7024 [ = Dessau 8722"] (Docimium in Phrygia) Pela(gius). 


Though it was not a frequent name, it was thus sufficiently common 
to negative the underlying idea of some investigations that it was 
specially invented to represent the native Celtic name of the most 
famous person who ever bore it. That he was far from being the 
earliest Latin bearer of the name is sufficiently proved by the fact 
that more than one Pope took it after his time. This clearly goes 
to show that the name enjoyed sufficient currency, apart from its 
evil association with the celebrated heresiarch, to make its survival 
possible among Christians. Finally, if there had been anything 
peculiar about the name, it is certain that Pelagius's adversaries 
would have made fun of it; but of this there is not a trace. 

Pelagius the Briton or the Irishman. Very little is known of 
the life of our Pelagius. The date and place of his birth and death 
are alike uncertain. His contemporaries are not even unanimous 
on the country of his origin. For, while Augustine S Orosius^, 
Marius Mercator^ and Prosper* speak of him as British (Britto, 
Br'itannus), Jerome appears to refer to him as Irish* {Scuttus). 
Professor Bury would reconcile the two statements by explaining 
that he was an Irishman born in Britain". According to Roman 
usage, Britto or Bntannus was applicable to any inhabitant of the 
Roman province or provinces named Britannia'^, and there is evi- 

1 Ep. 186 1 § 1 (C.S.E.L. Lvii 45) (a.d. 417?) Pelaginm, quern credijnus,utab alio 
distingueretur, qui Pelagius Terenti dicitiir, Brittoiiem fuisse cognomiiiatum, 

2 Apolog. 12 3 (C.S.E.L. v. 620) (a.d. 415) Britannicus noster. 

•' Liber Subnot. in Verba luliani prol. ed. Baluze (Paris, 1684) p. 2 (Migne, P.L. 
XLvni 111 a) (a.d. 431 — 2) Pelagium gente Britannum monachum. 

* Chron. a.d. 413 (Mommsen, Chronica Minora i 467) eodem tempore Pelagius 
Britto', carm. de ingrat. 1. 2 (Migne, P.L. li 94 b) (ca a.d. 430) coluber... Britanmis: 
so also in Gennadius, Uir. Inlust. 43 (ca a.d. 496) Pelagius Britto haeresiarches, 
but Richardson's edition omits Britto : Bed. H.E. i 10 (ca a.d. 731) Pelagius 

* If indeed the words in In Hierem. proph. prol. § 4 (C.S.E.L. lix p. 4 1. 20) 
(a.d. 415) stolidissimus et Scottorum pultibus praegrauatus, and In Hierem. m 1 § 4 
(C.S.E.L. Lix p. 151 11. 15 f.) habet enim progeniem Scotticae gentis de Brittannorum 
uicinia refer to him : they are for us of irritating vagueness. We now know that 
Alpinum (not Albinum) canem is the true reading (on p. 151 1. 13 of Reiter's ed. of 
Hier. in Hierem.), but these words could hardly refer to Pelagius. 

* Life of St Patrick (London, 1905) pp. 296, 43: 'It is possible that, as some 
claim, Pelagius was born in Ireland, but the evidence rather points to the conclusion 
that he belonged to an Irish family settled in western Britain': see sdso Hermathena 
xiii pp. 26 ff. 

■ Cf. W. M. Ramsay in Studia Biblica iv (Oxon. 1896) pp. 34 fif. 


dence of an Irish colony in Britain at an early date^ But the fact 
that the name 'Scottus' or 'Irish' undoubtedly conveyed a reproach 
in Jerome's day tells against the view that Pelagius was Irish. 
Claudian speaks of the Irish as cruel'^ and in several passages re- 
fers to Roman conflicts with them^. And Jerome himself, having 
learnt from historical or geographical works that the Irish had been 
guilty of a number of even worse crimes than that of cruelty, 
alluded to the Irish with contempt long before he ever heard of 
Pelagius^. Certainly the Irish of succeeding centuries, and few 
apart from them, treated Pelagius with the highest respect^. On 
the other hand, the literarystyleof Pelagius's work is so accomplished 
that it seems more probable that he obtained his higher education 
in Britain than in Ireland, which was never a part of the Roman 
Empire. The present writer is, however, more concerned with 
philological than historical problems, and does not feel called upon 
for a definite decision on this interesting question. 

His Commentary on thirteen Epistles of St Paul. Whatever his 
origin, Pelagius appears to have found his way to Rome towards 
the end of the fourth, or very early in the fifth century. We can 
only guess the purpose which took him there, or the source from 
which he drew the necessary means for travel, and support in the 
great city. The suggestion has been made that he had a serious 
disagreement with his father, and that he in consequence left home 
and lived away from Britain for the rest of his days^. Whether 
that conjecture be true or not, it is perfectly clear that he had 
received a first-rate education'', and it may thus be presumed that 

1 See Bury, op. cit. p. 350. 

2 De Bello Gothico 417 Scotto . . .truci. 

^ See the index in Koch's edition, ss.vv. Scotticus, Scottus. 

* Zimmer, Pelagius in Irlaiidp. 20, n.**, quotes them : Epist. 69 3 § 6 (U.S.E.L. 
Liv 684 11. 17 ff.) (ca A.D. 397) Scottorum et Aticottorum ritu...promiscuas uxores, 
communes liberos habeant; Adu. louin. ii 7 (Migne, P.L. xxiii 296 a) (ca a.d. 392) 
Scottorum natio uxores proprias non habet. 

^ The thesis of Zimmer's book, so ably defended. 

^ By the present writer in an article entitled, 'Pelagius' Doctrine in Relation to 
his early Life,' Expositor for 1915 vol. i pp. 180 ff. Other references bearing on the 
point, viz. in 1 Cor. iiii 16, xiii 4; Eph. vi 2, might be added to those given in that 

' H. Williams, Christianity in Early Britain (Oxford, 1912) p. 181, is in agree- 
ment ; see also pp. 199 f. 



he was of wealthy family. The records show^ that in Rome he lived 
a pure life in solitude, devoted to sacred study- His deep know- 
ledge of scripture will be sufficiently proved in subsequent parts of 
this book. He was also well read in the earlier Latin Christian 
writers as well as those of his own day. It has been generally 
believed that he had a first-hand acquaintance with various Greek 
Christian writers also. Certainly he shows on occasion the 
knowledge of classical Latin literature he had acquired in his 

Marius Mercator tells us that Pelagius ventured, before the 
devastation of the city of Rome-, to compose commentaries on the 
Apostle Paul and to circulate them among those on whose friend- 
ship he could rely. He believed himself to be explaining individual 
words or thoughts of the Apostle^. Augustine's language with re- 
gard to this work is very similar. He tells us in 412 that he 
had read certain writings of Pelagius (who was reported to him 
to be a holy man, far advanced in Christian graces), containing 
short expositions of the Epistles of Paul the Apostle*. These 
two writers not only allude to the commentaries, but, as will 
be set forth in detail later*, also make quotations from them. 
The book itself contains evidence that part of it at least was 
written not earlier than about 406 ^ Also the use of the 

1 On all these matters see chap. v. 

2 I.e. by Alaric the Goth in a.p. 410. 

* Commonitorium S7iper nomine Caelestii, ed. Baluze, p. 135 (Migne, P.L. xlviii 
83 a) (written in Greek a.d. 429, translated into Latin by the author, a.d. 431): 
ausus est memoratus (Pelagius) ante uastationem urbis Romae in apostohim Paulum 
commentarios condere, et hisedere, de quorum amicitiapraesumebat. Explanare autem 
se putauit xingula apostoli uerba uel sensus. The commentary is referred to also 
in the Liber Subnet atioiium, ed. Baluze, p. 2 (Migne, P.L. xlviii 111 — 112) (written 
A.D. 431-2). 

■* De Peccatorum Meritis et Remissione iii 1 § 1 (C.S.E.L. lx 129 11. 6ff.) legi 
Pelagii quaedam scripta,sanctiuiri,ut audio, et non paruo prouectuChristiani, quae 
in Pauli apostoli epistolas expositioties breuissimas continerent, 

^ Chap, ii pp. 35 ff. 

8 If we could tell the exact date of Jovinian's death, we could from this fact get 
a terminus post quern, for Jovinian was clearly dead at the time in Phil, iii 18 was 
written but all we can say is that Jovinian was certainly dead in 406 ; how much 
earlier than that we do not know (cf. W. Haller, Iovinianus...'Le\])zig, 1897 [— Te.rte 
u. Untersuchungen, N.F. ii Bd. Heft 2]) p. 131. Haller is absolutely ignoraut of 
the four references to Jovinian in our commentary. 


Origen-Rufinus commentary on Romans was not possible before 
about A.D. 405 ^ 

After the research of the past few centuries, there need be little 
hesitation in affirming that no copy of Pelagius's commentaries, 
with his name at the head of it, any longer exists. It is probable 
that originally this brief and unassuming work was issued without 
an author's name, and that, while in certain circles its authorship 
was known, and in others, where sympathy with its author's dis- 
tinctive views was felt, his name was actually attached to it, the 
majority of copies remained anonymous. From the time of Prae- 
destinatus- down to the middle of the seventh century, not a single 
quotation from the commentary by its author's name has been 
produced. But between the anonymous commentary on the Catholic 
Epistles^ of the latter date and the thirteenth century we find a 
considerable number of quotations from the commentary labelled 
' Pelagius,' as Zimmer was the first to point out*. Some of this 
evidence comes from glossed copies of St Paul's Epistles in Latin, 
and needs considerable sifting before it can be safely used. It 
seems, therefore, best to start merely with the quotations in 
Augustine and Marius Mercator, and to try to discover a commen- 
tary, which, though lacking, as it inevitably must, the name of 
Pelagius, shall be proved by internal evidence to be a reliable copy 
of the seemingly lost production. Before proceeding, however, to 
this identification, it is a duty to record the appearance of certain 
modern publications that have a bearing on our j)roblem. 

^ See chap, v below, pp. 188 f. 

2 I 88, a rather paraphrastic quotation from Pelagius's note on Eom. v 15 (cf. 
also Migne, P.L. liii 618 a, 619 a, 665 b). The much discussed Praedestinatus, written 
perhaps in the fourth decade of the fifth century, is probably the work of Aruobius 
Junior, cf. Dom Morin's Etudes, Textes, Decouvertes t. i (Maredsous and Paris, 
1913) pp. 315 ff., especially 345 n. 4; or Revue Benedictine, t. xxvin (1911) p. 158. 
The parallels with Sang. Aug. Merc, in this work are given by H. v. Schubert, Der 
sogenannte Praedestinatus (Leipzig, 1903) [=Texte u. Unters. N.F. ix (4)] pp. 33 ff. 

* This Irish-Latin commentary is contained in Cod. Aug. ccxxxiii (saec. ix) at 
Karlsruhe, and the reference to Pilag(ius) [in Eph. vi 14] sicut lurica — uirtutibus 
ornatur was first discovered by me. I am glad to accept Mr Esposito's date for it as 
the middle of the seventh century, half a century earlier than I ventured to put 
it in 1906 (Proc. Brit. Acad. vol. ii p. 431 [ = 23]) : see his careful and learned article 
in J.T.S. vol. XXI [1919—20] pp. 316 ff. 

* Pelagius in Irland pp. 162 f., summing up the first half of his book. 


History of Pseudo- Jerome Commentary in print. In the ninth 
and last volume of his edition of the works of St Jerome, published 
at Basle in 1516, Erasmus committed to the press a brief commen- 
tary on thirteen Epistles of St Paul, which he had found in a 
manuscript, bearing the name of Jerome as author. The words of 
the preface to'this part of his great edition are these : ' Postremo 
commentarios in omnes diui Pauli epistolas, quas Hieronymo 
uendicabat codex quidam, obsoletae uetustatis, Gotthicis characteHhus 
exaratus, sic sane perplexis et iam prae uetustate euanidis, ut coacti 
fuerimus in elementai-ior-um ordinem rursus descendere, et quod in 
ludo puelli faciunt, litterariis apiculis noscitandis operant dare: 
sed et ii sicut proxime superior es, indigni sunt iudicandi qui 
Hieronymo tribuantur. Qiiando enim sic ineptit Hieronymus, ut hie 
interpres quisquis is demum fiiit: quando sic balbutit, ut hie 
frequentissime soloecissat. Tametsi is qui glossam {ut uocant) 
ordinariam consarcinauit, Hieronymi titulo nonnulla citat, quae 
nominatim in iis conimentariis comperiuntur. At hoc neutiquani 
satis probauerit Hieronymi Stridonensis esse, quod scilicet sint illius 
nomine citata. Neque enim hoc agebat glossarius iste, svue Rahanus 
is fuit, siue alius quispiam, ut cuius assent inquireret, sed pro 
tempore quod ad rem suam faciebat, id excerpebat...quaedani 
inuenire licet, quae doctis etiam placere possint^.' 

No one has, I believe, ventured to dispute Erasmus' opinion, 
and to claim these commentaries for Jerome^. They have neverthe- 
less been reprinted in successive editions of the works of Jerome 
down to that in Migne's Patrology,in which, according to the example 
of Vallarsi, they get the very last place, just before the index. 
Some care seems to have been exercised in the reprinting of them 
in the editions of Erasmus, Victorius, and Martianay, but through 
sheer carelessness the text in Vallarsi (at least in the later quarto 
edition), and hence in Migne, is far inferior in accuracy to that of 
the editio princeps. To this subject we return later^. 

1 I have expanded the abbreviations of the original. 

- With the exception of a rather careless correspondent of Erasmus, Hieronymus 
Dunghersheym, writing from Leipzig to Erasmus on 18 Mar. 1517, who quotes 
from Ps. -Jerome on Phil, ii 6 (Opus Epistolarum Des. Erasmi Boterodaini, recogn. 
P. S. Allen, t. II (Oxon. 1910) no. 554 p. 507). 

3 Chap, vi pp. 281 f. 


Ambrosius Catharinus Politus, Archbishop of Compsa\ and 
Sixtus of Siena "^ observed that the commentaries were tainted with 
Pelagianism. The most important remarks of Sixtus are these : 
* Annotationes in tredecim Pauli Epistolas, quanquam in fronte 
praeferant praefatiunculam Hieronymi ad Heliodorum, scrihentis, 
styli tamen diuersitas, et dogmatam discrimen, ipsins non esse 
indicant. Ambrosius, Compsae Archiepiscopus, autumat, scriptoreni 
operis fuisse Pelagium, quia, in expositione octaui et noni capitis 
epistolae ad Romanos, doceat praedestinationem aeternam esse ex 
meritis electorum diuina praescientia praeuisis. Ego uero, ut libere 
pronunciein quid sentiam, nihil dubito authorein, quisquis illefuerit, 
Pelagiana peste laborasse, prout apertissime deprehendi potest ex 
annotationibus septimi capitis epistolae ad Roman, iibi edisserens 
illud, Sine lege peccatum erat mortuum, insanii^e^ eos putat, qui 
credunt peccatum Adae ex traduce parentum. in nos deriuasse, 
affi^rrnans, idem peccatum non propagations sed exemplo duntaxat 
ad posteros transfusum. Ex quo satis liquet, Hieronymum, Pelagiani 
dogmatis hostem acerrimum, non fuisse huius operis authorem. 
Sunt et alia quaedam his commentariis asserta, haereseos suspicione 
non carentia, quae nos una cum praedictis erroribus sequenti 
uolumine annotauimus\ Non sum tamen in sententia Ambrosii 
praeceptoris mei, ut arbitrer hos commentarios esse Pelagii: is enirn 
ut August, in 3. de baptismo paruulorum testatur, in explanationibus 
quas admodum breues edidit in epist. ad Roman, tanta modestia 
usus est, ut non sub propria, sed sub aliena persona proposuerit 
argunientum aduersus originate peccatum, quae August, ibidem 
ad uerbuni transcripta recenset.' In his own Annotationes^ also 
Sixtus takes occasion to criticize the pronouncements of the com- 
mentary on various subjects. In Ann. 238 (p. 516) he attacks the 
note on Rom. v 12 In quo onines peccauerunt, and quotes the pub- 

^' In omnes Divi Pauli... epistolas... comnientaria(Pskxis, 1566): 'quidam non in- 
curia Pelagium eius operis authorem credidere' (praef.) : quoted by Sixtus and Vic- 
torius, 11. ec. 

2 Bibliotheca Sancta (preface, 1566) (Paris, 1610) p. 247 (1) d, quoted below : 
(Colon. 1626), p. 309. 

3 We shall see afterwards that the passage here objected to is an interpolation 
into the original commentary. 

^ References to these follow immediately. 
5 =ed. Colon. 1626, p. 661 etc. 


lished notes on In hunc mundum peccatum intrauit et per peccatum 
mors: Exemplo uel forma. Et ita in omnes etc. (Migne 668 c) down 
to 'ipsa morietur': then In quo omnes peccauerimt, with the note 
on it'. Then follows the note on Rom vii 8, referred to above: 
'Item, si cum,' etc. (Migne 676 b) to 'non est legi subiectus' (676 c). 
He then points out that this error and interpretation are condemned 
by a decree of an African Council, cap. 77, with which the Synod of 
Trent later agreed. In Ann. 240 (p. 517) there is another reference 
to the note on Rom. %di 8. In Ann. 244 (p. 518) the note on 
Rom. viii 3 is criticized, because it insinuates that in Christ's 
flesh there was an inclination to sin : the words quoted are : ' Filius 
Dei, suscepta carne ' down to ' contagione peccati ' (Migne 679 c). 
In A7in. 251 (p. 522) the commentary is quoted among those which 
hold the view that predestination is after merits have been fore- 
seen : the particular passages mentioned are : Rom. viii 19 Prae- 
destinauit conformes fieri etc. down to ' conformes in gloria ' 
(Migne 685 a), ' quos praesciuit credituros ' — ' non inuitos ' {ibid.), 
Rom. ix 15 Miserehor, cui miserebor. ' Hoc recto sensu ' to ' sim 
misertus' (Migne 689 b). In Ann. 280^ (p. 537) the note on 
Gal, iv 24, Quae sunt per allegoriam dicta is referred to, the words 
cited being: 'Dedit regulam' to 'exponamus,' 'et hoc ipsum ibi' to 
'fieri uoluisse' (Migne 817 B, c). In Anii. 324 (p. 549) the note on 
1 Tim. iii 8 is quoted, namely the words ' Quaeritur cur de ' to 
' episcopos ordinauit ' (Migne 880 a, b). 

lo. Garetius of Louvain in his De Vera Praesentia Corpo7-is 
Christi in Sacramento EucharHstiae^, some years before Sixtus, had 
headed a section with the words ' Pelagius haeresiarcha, gi-atiam 
Dei oppugnare coepit circa annum 419,' and then proceeds to quote 
the notes, from that on 1 Cor. xi 23 Ego enim accepi a Domino 
(Migne 751 d) down to ' existamus ingrati ' (752 b), from Itaque 
quicumque manducauerit etc. to 'mundus manducabit' (752 c), 
from 'Unde oportet otiosum' to 'sancte percipiat,' and from Prohet 
autem se ipsum to 'habere, quam nouam' (752 d). He thus has not 
the slightest hesitation in regarding this commentary as the work 

1 There are minor differences from the printed text. Certain of the portions he 
objects to will afterwards be seen to be not by Pelagius. 
'■' The reference to this note in the index is false. 
^ Antv. 1561, class, viii p. 208 r. 


of Pelagius, and he makes in the margin a reference to 'B, August. 
lib. de peccatorum meritis et remissione ' in connexion with it. 

Similar statements were made by the second great editor of 
Jerome, namely Marianus Victorius of Reate, Bishop of Amei'ia, 
who writes as follows : ' Commentarii in tredecim Paidi Apostoli 
epistolas, Hieronymi, ipsi etiam minime stmt: nam nee eius phrasim 
habent, nee eruditionem : quamuis in f route operis epist. ad Helio- 
dorum inscriptam, contineant: Quin potius hominis esse Pelagiana 
labe conimacidati compertissinium est (ni ea tantum quae orthodoxa 
non sunt, ah haereticis potius sint adiecta). Notauit hoc ante nos 
Amhrosius Cathariniis Politus Compsae Archiepiscopus, catholicae 
religio7iis assertor: notauit et Sixtus Senensis, quarto et sexto Bib- 
liothecae sanctae uolumine: uterque errores Pelagianae haereseos, 
qui eo in opere inaeniuntur, redarguentes. Si quis autem scire hoc 
concupiscat, liquido ex expositione quinti ad Romanos capitis inueniet: 
ubi Adae peccatum in humanum genus, non propagatione, et traduce, 
sed exeniplo tantum, et imitatione diffundi asserit: ac si illud non 
contrahant, nisi qui aetate adulta, et ad peccandum iam apta, Adam 
sceleribus imitentur. Quod ipsuni ex interpretatione septimi capitis 
liquidius apparet. Enarrans enim illud. Sine lege peccatvm erat 
MORTVVM, ita scr'ibit. Itera^ si cum lex non esset, peccatum mortuum 
est: insaniunt qui de Adam per traducem asserunt ad nos venire 
peccatum. Quae opinio quam impia, et haeretica sit, nemo catholi- 
corum ignorat. Ego quid de operis auctore statuam, certi habeo 
nihil: nisi quod Hieronymi illud non esse, compertissinium est. Edi 
tamen curauimus eo modo, quo antea typis mandatum circumfere- 
batur, ne quid priori editioni deesset, nulla quideni syllaba a nobis 
auctum, diminutum, aut immutatum: Lectorum arbitrio, quid de 
illo eodstimare uelint, penitus derelinquentes"^! 

The position of Robert, Cardinal Bellarmine, is the same, as 
appears from these words: ' Conimentaria in omnes Epistolas, non 
solum non sunt Hieronymi, sed, quod magis mirum est, Pelagii 
haeresiarchae sunt. Nam sanctus Augustinus, lib. 3. de peccatorum 
meritis, et remiss, cap. 1. dicit, se legisse Gommentaria Pelagii breuia 

1 The words Item to uenire peccatum are interpolated, as we have seen, p. 7 n. 3. 

- Sancti Hieronymi Stridonensis Operum Tomus iiii (Paris, 1609) pp. 461 — 462; 
the date of the eighth volume of Victorius' earliest edition is 1572 (Schoenemann, 
Biblioth....Patrum Latinorum, t. i [Lips. 1792] p. 503). 


in omnes Epistolas Pauli: et infra cap. 12. allegat quaedam ex 
Commentario in caput 7. prioris ad Corinth, quae in his Commen- 
tariis inueniuntur. Vide auctorem horum Commentarionnn, in cap. 
5. 6. 7. et 8. et 11. ad Romanos; item 1. Corint. 4, Philip. 1. c& 3. 
t& in caput 6. prioris ad Timot. et inuenies sententias manifeste 

G. J. Voss, in one of the most learned books ever published on 
the Pelagian controversy^, sums up his own and his predecessors' 
research thus : 'Etiani Pelagius edidit xiv^ Expositionum lihros in 
epistolas Paulinas. Meminit eorum Augustinus lib. ill. de peccat. 
meritis et remiss, de gestis Palaestinis cap. xvi. Valde autem 
eo inclinat animus, ut putem eos ipsos esse, qui commentariis Hiero- 
nynii in epistolas aliquot Paulinas subjungi vulgo solent. Qui non 
tenui suspicioni locum magnum dedit, quod scriptor hie statuat passim, 
neminem per Adamuni peccati reum esse, nisi quatenus eius imitetur 
exemplum: cumque omnes in eo peccasse dicuntur,id non d7r\(io<;in- 
telligi dehere, sed de genere improborum: quando alii praeter hos 
sint, qui ju^te vivendo secundum Adamum, hoc est Christum, sequi 
anient. Audiamus ipsum in cap. v. ad Rom. sic scribentem : then 
follow "Ut qui sequentes" — "reconciliaremur Deo" (Migne 668 b). 
Et mox: "Quomodo cum non esset" — "est revocata" (ibid.). Ibidem 
ad illud, Et ita in omnes homines (Migne 668 c) — "ritu vivebant" 
(ibid.). Et pauculis inter jectis : In quo omnes peceaverunt: "hoc 
est" (ibid.) — "exemplo Adae peccant" (668 D). Et mox: "Forma 
Christi Adam factus^" (Migne 669 b) — "eum volentibus^" (669 c). 
Et post aliqua ad ilia verba : Sicut per unius delictum in omnes, 
etc. "Quomodo," inquit, "potest unius" (Migne 670 c) — "significat 
multitudinem" (671 A). Et mox: "Sicut exemplo inobedientiae " 
— "j ustificantur multi*" (ibid.). Item in cap. vi. "Qui veterem ho- 

^ De Scriptoribus Ecclesiasticis Liber Unus (Col. Agr. 1613) p. 165, or ed. 6 (Lovan. 
1678) p. 120 = Opera, t. vii (Colon. 1617) p. 73 b— d. Ph. Labbe, De Scriptoribus 
Ecclesiasticis t. i (Paris, 1660) p. 441, objects to the Pelagian authorship on the 
ground of the absence of the Augustinian quotations from the published commentary. 

- Historiae de Controversiis, qiias Pelagius eiusque reliquiae rnoueritnt, libri 
septem (2nd edition) (Amstel. 1655, sm. 4to) pp. 11 if. (original ed. 1618). 

3 An error due to carelessness, found also in other works on the subject. 

^ We shall see that this passage is an interpolation on the original Pelagius. 

* So Voss : Migne cupientibus. 

* Here Migne is corrupted by homoeoteleuton : see my text in vol. ii : Voss is 
using an earlier and better edition. 


minem"(Migne 672b) — "imitando peccabant" (ibid.). Item: "Homo 
membra sua "(Migne 673 a) — "arbitrii libertatem"(ibid.). Ubipalam 
est, eurti naturam liberi arhitrii considerare citra ullaru Spiritus S. 
gratiam. Quemadmodum et cum paidlo post ad illud, Humanum 
dico, propter etc. "Humana," inquit, "ratione" — "nullus abnuere" 
(Migne 674 a). Sed omnium maxime scriptorem prodit, quod cap. 
sequentijioc est septimo, ait: "Insaniunt qui" — "venire peccatum" 
(Migne 676 b)\ Idem in epist. ad Philipp. cap. ii.^ "Occasio fidei" 
— "utique crederemus" (Migne 844 b). Paullum hie pedem. figamus. 
Agnoscit honitatis esse divinae, quod Christus venerit in miindum, ut 
nos doceret, quid credere debeamus: absque hoc esset, credenda ig- 
norarentur, eoque nee crederentur. Quid uero aliud hie requirit, 
quam ut bonitas divina nobis doctores mittat, qui extrinsecus instruant 
verbo ? Nam opem spiritus Sancti ad fidem ingenerandam nee hie 
necessai'iam esse dicit, nee alibi usquam. Unde cognoseimus, quomodo 
intelligere oporteat, quod in cap. XI. ad Rom. ait ; Fidem bonitate 
Dei conferri (Migne 698 b). Cujusmodi a Catholicis omnibus pro- 
barentur, nisi constaret a quibus dicerentur. Atque ut haec ejus libri 
scriptorem Pelagianum esse dare osteiidunt, ita eum non alium quam 
ipsum Pelagium esse verisimile faciunt, quae de Pelagii libro refert 
Augustinus. Nam lib. in. de peeeatorum meritis et remissione cap. i. 
seribit in hunc modum, Post paucissimos dies legi Pelagii quaedam 
scripta, quae in Pauli Apostoli epistolas expositiones brevissimas 
continerent. At conimentarii quos habemus, sunt brevia in omnes 
Pauli epistolas scholia. Subjicit mox Augustinus: Atque ibi com- 
peri, cum ad ilium venisset locum, ubi dicit Apostolus : Per unum 
hominem peccatum intrasse in mundum, et per peccatum mortem, 
atque ita in omnes homines pertransisse : quandam eorum argu- 
mentationem, qui negant parvulos peccatum originale gestare. 
Argumentatio ilia sic posita est ; Si Adae peccatum etiam non 
peccantibus nocuit, ergo et Christi justitia etiam non credentibus 
prodest. Atqui hunc in modum, ratiocinatur, qui commentarios istos 
scripsit: quemadmodum videre ex iis fuit, quae supra adduximus. 
Quanquam ilia ipsa penitius eo7isideranti fatendum est, satis obscure 
hanc ex iis sententiam elici. Verba sane ipsa non legas. Sed fortasse 
haec inde exeiderint: quod arbitratur Gretzerus defensione Bellarmini 

1 On this passage see p. 7 n. 3. 
^ This should be i. 


de verho Dei lib. iv. cap. v. Praeterea cap. v. ejusdem lihri ait 
Augustinus, Pelagiuni non uno niodo explicuisse illud Pauli ad 
Rom. V. Adam forma futuri. Atqui hoc facit commentator iste, his 
verbis: "Sive ideo forma fuit" (Migne 669 b) — "Sicut enim Adam 
etc.^" (ibid.). Denique apud Augustinum duodecimo ejusdem, libri 
capite (ubi Paidinum illud expendit 1. Cor. vii. Sanctificatus est 
vir infidelis in uxore, et sanctificata est mulier infidelis in fratre : 
alioqui filii vestri immundi essent, nunc autem sancti sunt) magnus 
haec commentatur antistes: Aut sic est accipiendum, quemadmodum 
et nos alibi, et Pelagius, cum eandem ad Corinthios epistolam 
tractaret, exposuit, quod exempla jam praecesserant et virorum, 
quos uxores, et feminarum, quas mariti lucrifecerant Christo, et 
parvulorum, ad quos faciendos Christianos voluntas Christiana etiam 
unius parentis evicerat. Atque ita interpretatur scnptor horurn 
scholiorum. Verba eius apponam: "Exemplum refert" (Migne 736 d) 
— " verbo lucrifiant"(737 a). Pelagium igitur istorimi esse scholiorum 
scriptorem, verisimillimum, videtur. Interim de hoc nolumiis cum 
quoquam ducere contentionis funem: dummodo si non Pelagii, saltern 
Pelagiani alicujus credantar: quod etiam agnoscunt Sixtus Senensis 
lib. IV. Bibl. Sanct. verbo Uieronymus : Johannes Garetius de sacri- 
ficio Missae centur. Y.fol. 43. Bellarminus in catal. script. Eccles. 
et lib. IV. de verbo Dei cap. v. et lib. iv. de amiss, gratiae, atque alibi: 
item Pererius comm. in Gen. cap. xiv, n. 8. Marianus Victorius 
censurd tom. 8. Hieron. Scultingius in censurd confess. Hieron. 
Salmeron in epist. Pauli disjx xvi. Coccius Thesauri T. 2. artic. 11. 
ut interim de Witakero, Perkinso, lunio, Rob. Coco, pluHmisque 
aliis nihil dicam^.' 

To recapitulate. Erasmus had contented himself with denying 
the paternity of Jerome. Succeeding scholars pointed out the 
Pelagianism in the commentary, and some boldly identified it with 
the commentary of Pelagius himself mentioned and quoted by 
Augustine. The difficulty that certain of Augustine's quotations 
are wanting to the would-be Jerome, was hardly faced by anyone. 
With the publication of the first printed edition of Marius Mercator 

1 This last part is an interpolation in Pelagius. , 

2 Of the authorities here enumerated I have consulted such as were accessible to 
me : some are quoted above : the reference to Pererius appears to be wrong. On 
Phil. Labbe's view, published in 1660, see above, p. 10 n. 1. 


in 1671, and particularly that by Jean Gamier in 1673, the problem 
assumed a somewhat different aspect. 

Slightly prior to Garnier, however, Enrico de Noris, of Verona, 
had observed the bearing of the new information supplied by 
Mercator on the subject of the Pelagian commentary, but it did 
not lie within his province as a historian to work the matter out 
with Garnier's fullness \ It does not appear that either author's 
work was known to the other, and we have here therefore an example 
of that curious coincidence in research which occurs so often. Noris 
alleges a reference to Pelagius's commentary in cap. 16 of Augustine's 
De Peccato Originali, which, so far as I know, had not been previously 
observed^ and he quotes Mercator on the date of the Pelagius 
commentary, direct from the Vatican codex, p. 37^. He then 
mentions Jansen's agreement with Voss that the Hieronymian 
commentary is undoubtedly that by Pelagius, and proceeds to con- 
firm this view by the quotations Marius Mercator gives from the 
commentary*. An interesting suggestion which he then makes must 
be quoted in his own words: ' Scio S. Doctorem [i.e. Augustine] 
lib. 3. de pec. mer. et remis. nonnullas ex eisdem Pelagii commentariis 
argumentationes ad verbum recitai'e, quae in editis non leguntur, sed 
illud satis verisimillimum judico, eas ratiocinationes Celestii manu 
insertas, eoque pacto in S. Doctoris mantis venisse, etenim familiare 
Celestio fuit, brevissimis syllogismis ad haeresim stabiliendam nti; 
Jansenius existimavit, easdem expositiones Pelagii ipsius manufuisse 
subinde pariimper immutatas, de quo nolo cum eodem ducere con- 
tentionis funem. Porro apertis verbis ibidem peccatum originale 
excluditur,namcap.5.ista leguntur: "Quomodo potest" (Migne670c) 
— "homines iustificati sint^?" (671 a) quo loco pluribus tradit, Adam 
nocuisse poster'is exemplo tantu7n, ac forma, qua mors animae, non 

^ Historia Pelagiana et Dissertatio de Synodo V. Oecumenica, etc. (Patav. 
1673: also Amstel. 1677). The dedication is dated 23 March 1673, but the permis- 
sion of the Prior General of the Augustinians is dated 20 June 1672 ; the book 
must therefore have been completed before the latter date. For the use of Mercator 
by De Noris cf. pp. 5, 14 f. etc. of ed. Amstel. 

^ p. 14 ed. Amstel. There is a clearer instance, however, in c. 21 § 24, which he 
failed to quote: unless indeed 16 be a wrong reference, for 21. 

^ p. 14 ed. Amstel. 

* For these, see below, pp. 41 ff. 

' Migne sunt. 


vero corporis Adami culpam imitantibus intimatur. Et cap. 7. totum 
se prodit inquiens: "Insaniunt" — "venire peccatum" (676 b). Vides 
ergo Pelagium strenuum Rufini defensorem. Praeterea in eisdem 
commentariis iiniversum haeresis, quam postea fusius docuit, virus 
evomuit. In c. 8 laudatae epistolae ad Romanos ad ilia verba: Lex 
enim — "appellat" (Migne 679 b), et inferius: "Quicumque secundum 
doctrinam" — "Deiaguntur^" (Migne 681 d); Neutrobique enim gra- 
tiam Spiritus Sancti a lege, et doctrina distinguit. In cap. 11. vers. 7. 
Electio autem consecuta est, haec scribit: "Divina Scriptura" (Migne 
696 a) — "dedit libertatem " (Migne 696 b). Quibus actuate Dei 
auxilium prorsus negavit, admittens tantum adjutorium possibilitatis, 
quod idem est, ac liberum arbitHum naturae rationali insertum. 
Augiae^ stabulum purgare mallem, quam illorum commentariorum 
errores in indicem cogere^.' 

Garnier did not content himself with the issue of Mercator's 
text, but furnished it with copious notes and dissertations, which 
still possess interest and value ^ He mentions that some attributed 
the Hieronymian commentary to Priraasius, others to Sedulius' : 
but of these attributions I can find no trace. That there is, however, 
a close relationship both between 'Primasius' and 'Hieronymus,' 
and between Sedulius and 'Hieronymus' we shall afterwards see^ 
Garnier himself states that in his time no one doubted that the 
commentaries were not merely by a Pelagian, but by Pelagius 
himself He then quotes Praedestinatus and the passages of 
Augustine, to which reference has already been made, including 
Be Peccat. Orig. c. 21'' and Op. Imp. c. Iidian. i c. 54, which last 
passage, so far as I know, had never been adduced by anyone else^ 
As has been said above, he also takes full advantage of the evidence 
of Mercator in support of the Pelagian authorship. But perhaps 

1 An interpolated passage. * Sic. 

•■' Ed. Amstel. p. 15. 

■* Marii Mercatoris S. Augustino aequalis Opera quaecumqne exstant. Prodeunt 
nunc primum studio J. Garnerii: reprinted in Migne, P.L. xlviii. The 'nunc pri- 
mum' is not strictly correct, as Schoenemann, Bibliotheca Patrum, t. n (Lips. 
1794) p. 550 = Migne, P.L. xlviii 53 a — b, shows : an earlier edition appeared in 1671. 

5 Migne xlviii 83 b. 

6 Chap. vipp. 322 ff., 338 f. 

7 See above, p. 13 n. 2. 

^ Migne xlviii S4a — b: Labbe, p. 795, had noted Sedulius's connexion. 


his greatest service in connexion with the whole matter was to call 
attention to a passage in Cassiodorus that had been overlooked by 
all the earlier investigators ^ 

Cassiodorus, in the eighth chapter of his invaluable work, 
Institutiones Divinarum Litterai'um, written about the middle of 
the sixth century, is describing Latin commentaries on the New 
Testament Epistles contained in the library of the monastery he 
had founded at Vivarium. This chapter is of the utmost importance 
to its subject, and must be quoted here, because of the light it 
sheds on the problems with which we are dealing. By the kindness 
of Dr C. H. Turner, I am enabled to give the text according to the 
Bamberg eighth century MS, the best of all ^ 

§ 1. Octauus codex canonicas epistulas continet apo.stolorum. sed in epi- 
stolis tredecim sancti Pauli annotationes conscriptas in ipso initio meae lecti- 
onis inueni, quae in cunctorum manibus ita celebres habebantur, ut eas a 
sancto Gelasio papa urbis Romae doctissimi uiri studio dicerent fuisse con- 
scriptas. quod Solent facere, qui res uitiosascupiunt gloriosi nominisauctoritate 
defendere : sed nobis ex praecedentibus lectionibus diligenti retractatione 
patuerunt subtilissimas quidem esse ac breuissimas dietiones, sed Pelagiaiii 
erroris uenena illic esse seminata. et ut procul a uobis fieret error hereticus, 
primam epistolam ad Romanos qua potui curiositate purgaui, reliquas in 
chartaceo codice conscriptas uobis emendandas reliqui : quod facile subiacebit, 
quando praecedenti exemplo audacior reddatur sequentis imitatio. 

§ 2. Sed inter has sollicitudines grauiter aestuatus, quendam anonymum 
codicem subnotatum diuina reperi prouisione coUatum, qui tredecim ej^istulas 
sancti Pauli non ignobili annotatione tractauit. hie diligenter excussus 
secundum uobis ac securum genus commentorum, domino largiente, prae- 

§ 3. Ad Hebreos uero epistulam, quam sanctus lohannes Constantino- 
politanus episcopus triginta quatuor omeliis Attico sermone tractauit, 

1 Migne 84 c. 

2 Ed. Garet, t. ii (Venet. 1729) p. 514, Migne lxx p. 1119 : the Bamberg MS 
is HJ iv 15 (Patr. 61), saec. viii ex.; see E. A. Lowe, The Beneventan Script 
(Oxford, 1914) passim. It appears to be a direct, or almost direct, copy of the 
Cassiodorian autograph. P. Corssen, Jahrbilcher fur protestantische Theologie ix 
(1883) pp. 619 — 633, made admirable use of another part of it, as also Th. Zahn, 
Gesch. d. ntl. Kaimis ii Bd. (1) (Erl. u. Leipz. 1890) pp. 268 ff. Another really old, but 
fragmentary, MS of this work is at Vercelli, Bibl. Capitolare, clxxxiii (saec. viii), 
of which one page is photographed in Specimina Codicum Latinorum Vatieanorum 
coll. F. Ehrle et P. Liebaert (Bonn, etc. 1912) no. 9. There are MSS also at Here- 
ford, Karlsruhe, St Gall, etc. 


Mucianuin uirum disertissimum transferre fecimus in Latinum, ne Epistu- 
larum ordo continuus indecoro tennino subito rumperetur. 

[The next four paragraphs are omitted here, becaitse they concern only the 
Canonical or Catholic Epistles.] 

§ 8. Tertium uero codicem reperi epistolarum sancti Pauli, qui a nouniillis 
beati Hieronymi annotatioues breuissimas dicitur contiuere, quern uobis 
pariter, Christo largiente, dereliqui. 

§ 9. Post haec uero tria paria, quae diximus, commentoriun, Petrus, abbas 
Tripolitanae prouinciae, sancti Pauli epistulas exemplis opusculorum beati 
Augustini subnotasse narratur, ut per os alienum sui cordis declararet 
arcanum ; quae ita locis singulis competenter aptauit, ut hoc magis studio 
beati Augustini credas esse perfectum. mirum est enim sic alterum ex altero 
dilucidasse, ut nulla uerborum suorum adiectione permixta desiderium cordis 
proprii cotnplesse uideatur ; qui uobis inter alios codices, diuina gratia suffra- 
gante, de Africana parte mittendus est^ 

§ 10. Sic totus ordo epistularum canonicarum tarn sancti Pauli quam 
diuersorum apostolorum domini fauore completus est. 

§ 11. Dicitiu" euim et beatum Ambrosium subnotatum codicem epistu- 
larum omnium sancti Pauli reliquisse, suauissima expositione conpletum, 
quem tamen adhuc inuenire non potui, sad diligenti cura perquiro. 

At this point we may leave the text of Cassiodorus, following the 
example of the men of that day. We shall see later that if they 
had read farther, they would have solved one problem at once. 

Garnier's interpretation of Cassiodorus' statements is this. 
After paraphrasing § 1, he expresses doubt whether Cassiodorus's 
pupils ever carried out the revision he suggested, but points out 
that in published editions of the Hieronymian commentary the 
passages alleged by Augustine, Marius Mercator, and Praedesti- 
natus as tainted with Pelagian error, are wanting : the probable 
reason for this is that our copies are descended from the copy made 
by Cassiodorus for his monks. Gamier feels the difficulty of the 
presence of the Insaniiint passage in Rom. vii 9 (Migne 676 b), and 
explains that, after it had been removed by Cassiodorus, it had been 
inserted again by a scribe 2. In a later part of his book he subjects 
the whole question to a somewhat more elaborate examination, and 
sets forth the quotations from Pelagius in Mercator side by side 

1 This work has not been printed, but a MS exists in the Vatican, lat. 4950 
(saec. xi), formerly S. Petri Damiani in AveUino, as Dr H. M. Bannister informed 
me. See also Denifle, Luther und Luthertum i (2), p. 22. 

- As a matter of fact, this passage was never before Cassiodorus's eyes, cf. p. 7 n. 3. 


with the corresponding Hieronymian comments. His conclusion is 
that the would-be Jerome is the original . Pelagius as revised by 
Cassiodorus, who made excisions in the interests of orthodoxy ^ 

The industry which Garnier displayed gained a signal reward: 
his opinion remained practically undisputed for two centuries. 
Succeeding statements are quoted here for the sake of relative 
completeness, but to all intents and purposes Garnier's view held 
the field right down to and even beyond the time of Klasen (1885). 

Cave^ obviously depends in part on Garnier. He mentions the 
evidence in Augustine, Marius Mercator, Praedestinatas, and 
Cassiodorus, that Pelagius wrote a commentary on ' xiv ' Epistles 
of St Paul, and that Vossius and most scholars regard it as identical 
with the Hieronymian commentary in our possession. He records, 
however, at some length Ussher's dissent from this opinion ^ 
Ussher's views show an interesting originality. He regards Pelagius 
as the author of the first commentary mentioned by Cassiodorus, 
but considers that the Hieronymian commentary is a comparatively 
late compilation from Pelagius, Jerome, and Primasius, a jumble 
of contrary opinions, heterodox from the first of these sources, 
orthodox from the second and third, put together by an ignorant 
person. He cites a contradiction between in Rom. xi 22 that 
'faith' is conferred by God's goodness (Migne 698 b) with many 
other passages, on the one hand^, and Pelagian views on the other, 
and holds that therefore those who attribute the whole work to 
Pelagius are mistaken. He then calls attention to the fact that 
three passages quoted by Augustine from Pelagius, are nowhere to 
be found in the Hieronymian commentary, and that the Pelagian 
prefaces and arguments to the Epistles are also absent from itl 
He allows, however, the presence of Pelagian material in the com- 

1 Diss. VI pars ii cap. 2, 'Expositiones breves in omnes Epistolas Pauli apos- 
toli' (Migne xlviii 587 c — 593 a). 

- ScriptorumEccleslasticorum Historia Literaria {hond. 1688) i)Tp. 292 f. : (Colon. 
1720) pp. 244 f. 

^ Abp Ussher p. 574 (of what work I cannot discover, prob. MS, see p. xlvi of 
later ed. of Cave). 

■* He quite rightly says that no other author more frequently insists on our justi- 
fication by the free grace of God, and faith alone. See my chap, iii p. 70. 

■5 This is the first occasion on which that element enters into the controversy ; 
Ussher of course knew the Book of Armagh, where this prefatory material occurs 
under the name of Pelagius : see below, p. 25. 

S. P. , 2 


mentary, namely the already quoted note on Rom. vii (Migne 
676 B)',the remaining quotations in Augustine, and the quotations 
made in the Irish Canons-. The best part of it, however, he thinks 
was taken from the very brief notes of Jerome, of which Cassiodorus 
speaks, as is suggested by the name of Jerome in the title, and the 
preface to Heliodorus. He reminds us that Walahfrid Strabus in 
the Glossa Ordinaria constantly quotes, and Sedulius^ once {in 
1 Cor. vii. Migne cm, 144 a) appears to allude to the com- 
mentary, under the name of Jerome ; ' non parentes ' (Migne 
740 a) — 'incorruptam seruare ' (740 b)*. Cave concludes by 
quoting Garnier's opinion as to the Hieronymian commentary in 
its present state. 

Du Pin in his Nouvelle Bihliotheque des Auteurs Ecclesiastiques^ 
says: 'Les Commentaires ou les notes sur toutes les Epitres de 
Saint Paul ne sont point de Saint Jerome, mais d'un Auteur 
Pelagien,qui enseigne ouvertement ses erreurs en plusieurs endroits, 
et principalement sur le chapitre 7. de I'Epitre aux Romains. II 
est certain que Pelage avoit fait un Commentaire sur les Epitres 
de Saint Paul, que Saint Augustin cite en qiielques endroits du 
troisieme livre des merites et de la remission des pechez. Ce meme 
Commentaire de Pelage est encore cite par Marius Mercator, et Ton 
trouve dans celui-cy la pluspart des passages rapportez par ces deux 
Auteurs. II yen a neanmoins un ou deux qui ne s'y trouvent pas; 
ce qui donneroit lieu de douter si c'est entierement le meme, si 
Cassiodore ne nous avertissoit qu'il en a retranche quelques 
passages.' In a later note" he adds: 'Quelques-uns doutent si ce 
Commentaire est celui que Saint Augustin cite sous le nom de 
Pelage: 1. parce qu'on trouve aussi par mi les CEuvres de Saint 
Ambroise un Commentaire Pelagien sur les Epitres de Saint Paul: 
2. parce que Ton ne trouve pas tous les passages que Saint 
Augustin rapporte, comrae etant du Commentaire de Pelage, ou 
du moins ne les y trouve-t-on pas dans les memes termes. La 

^ pp. 7 ff. - The reference to the Irish Canons first appears in Ussher. 

3 The Bamberg MS of Sedulius here assigns to H., but the reference may be 
to some passage of the genuine Jerome, perhaps Adu. louin. i 13 (cf. J.T.S. 
vol. xviii p. 228). 

■» This passage is an interpolation ou the original Pelagius. 

5 T. Ill (Paris, 1689} p. 426. 

6 Op. cit. p. 492 note c. 


premiere de ces deux raisons est tres-foible, puisqu'il est fort 
possible qu'un Auteur Pelagien ait fait des Commentaires sur 
Saint Paul, differens de ceux de Pelage, La seconde seroit de 
quelque poids, si Ton ne trouvoit pas dans ce Commentaire attribue 
a Saint Jerome la pluspart des passages citez par Saint Augustin. 
Car 1. Saint Augustin dans le ch. 16 du livre des Actes de Pelage 
dit, que cet Heretique a explique ces paroles du chapitre 9. de 
I'Epitre aux Remains, Neque volentis, neque currentis est Dei (sic !), 
en disant que Saint Paul avoit ainsi parle par interrogation, Voce 
interrogantis et redarguentis. Cette meme explication, ces memes 
mots se trouvent dans le Commentaire dont nous parlons. 2. Saint 
Augustin au livre 3. des Merites des pechez chapitre 12. dit, que 
, Pelage expliquant ce passage du chapitre 7. de I'Epitre aux 
Corinthiens, Sanctijicatus est vir infidelis, remarque qu'il y avoit eu 
des exemples de femmes Fideles qui avoient converti leurs maris 
Infideles. Cette meme remarque est dans ce Commentaire. 3. Saint 
Augustin dans le meme livre chapitre 4. dit, que Pelage a dit sur 
ces paroles Rom. 5. Quae est forma futuri, qu'elles se peuvent 
entendre de plusieurs manieres. La meme chose est remarquee 
dans ce Commentaire; mais ce qui met la chose hors de doute, 
c'est que Marius Mercator dans son Memoire instructif cite un long 
passage tire des Commentaires de Pelage, qui se trouve tout entier 
dans celui-ci. II est vrai que S. Augustin au livre 3. des Merites 
des pechez (sic!) chapitre 2. rapporte un argument centre le peche 
originel, qui ne se trouve point dans ce Commentaire, et qu'il cite 
dans le chapitre 3. un endroit qui est aussi rapporte par Marius 
Mercator, qui n'est point non plus dans ce Commentaire attribue 
a Saint Jer6me. Mais il y a apparence que ces endroits ont et^ 
effacez et raiez par quelques Catholiques.' 

Richard Simon, the greatest of all students of ancient com- 
mentaries on the New Testament, records the opinions of Sixtus 
of Siena, Catarinus, Bellarmine and Labbe\ Without naming 
Garnier, he adopts the same opinion as his, that the Hieronymian 
commentary is the work of Pelagius, as revised by Cassiodorus. 
He makes a curious mistake, in stating that Cassiodorus reports 
that some assigned the commentary with the Pelagian poison to 

1 Histoire Critique des principaux Commentateurs du Nouveau Testament... 
(Rotterdam, 1693) c. 16 pp. 236 ff. 



Primasius^ Cassiodorus, as a matter of fact, never mentions 
Primasius in this connexion. But in spite of such a mistake as this, 
Simon's chapter on the Pelagian commentary is, as might be 
expected, one of the most valuable ever written. We shall return 
later to his work, but at this stage it may be well to quote some of 
his further remarks. 'Pelage fait paroitre dans tout cet ouvrage, 
qui ressemble plus a des Scolies qua un Commentaire, qu'il etoit 
exerce dans le stile des Livres Sacrez. Si Ton excepte quelques 
endroits de S. Paul, qu'il a interpretez selon ses faux prejugez, il 
peut etre mis au rang des habiles Commentateurs du N. Testament. 
II paroit meme que Sedulius, Primasius, Haimo, et quelques autres 
en ont copie la meilleure partie dans leurs Commentaires. Bien 
qu'il n'ait point expliqu^ I'Epitre aux Ebreux, il la cite souvent 
sous le nom de Saint Paul. II a eu apparemment egard a I'usage 
de quelques Eglises Latines, qui ne la lisoient point alors dans les 
assemblees publiques....Si Ton ote ce qui regarde la grace, et 
quelques autres points qui sont connus de tout le monde, Pelage a 
suivi la creance commune de I'Eglise, qu'il a meme defendue avec 
vigueur contre les Heretiques. Son Commentaire sur S. Paul 
merite d'etre lu, mettant a part ses erreurs....le nom de Pelage qui 
est devenu odieux, ne nous doit point detourner de la lecture de 
son Commentaire, ou il donne des preuves de son habilete....L'on 
prendra meme garde, que pour ne pas s'accorder avec la doctrine, 
qui a ete la plus commune apres S. Augustin parmi les Latins, il 
n'est pas pour cela Heretique, dans tons les endroits ou il ne con- 
vient point avec ce Pere: autrement il faudroit accuser d'heresie 
la pliipart des anciens Docteurs de I'Eglise. Je croy meme que 
Pelage avoit compose son Commentaire sur les Epitres de Saint 
Paul, avant qu'il eut ete declare novateur. Comme Ton est oblige 
de rendre justice a tout le monde. Ton distinguera ce qu'il a de 
commun avec les anciens Ecrivains Ecclesiastiques, d'avec ce qu'il 
a avance de luy-meme, sans etre fonde sur la Tradition. Car c'est 
en cela seulement qu'on peut I'accuser d'avoir ete novateur.' The 
justice of these remarks is evident, and it is regrettable that they 
required to be made. One is also impressed by the fact that Simon 

1 It looks as if Simon here had unwittingly taken Garnier's note on Marius 
Mercator for a quotation from Cassiodorus himself. On p. 238 he also quotes 
Cassiodorus inexactly. 


had read through the published commentary with care. I take leave 
to doubt whether many of the others who have written about it, 
have done the same. Otherwise, it is strange that for four centuries 
the text should have been suffered to remain so ridiculously 

The great Tillemont's contribution to this question is in these 
terms: 'Nous avons parmi ses oeuvres (i.e. those of Jerome) un 
commentaire sur toutes les epistres de S. Paul, hormis sur celle 
aux hebreux. Mais quoiqu'il porte le nom de ce Saint, et qu'on 
y trouve a la teste une lettre a Heliodore, neanmoins et les 
Catholiques et les Calvinistes conviennent qu'il n'est point de Saint 
Jerome, raais de quelque Pelagien, et apparemment de Pelage 
mesme, [ce que nous n'examinerons pas ici. C'est peutestre celui] 
dont on avoit parle a Cassiodore; car il ne dit point que I'epistre 
aux Hebreux y fust. Mais il ne le dit point non plus de celle aux 
Remains, que Saint Jerome n'avoit point encore expliquee, lorsqu'il 
ecrivit a Alga^e [en Fan 407 ; et il ne paroist point qu'il I'ait jamais 
fait]. Cassiodore parle encore de quelques notes fort courtes sur les 
epistres de Saint Paul, que quelques uns disoient estre de S. Jerome. 
[Je n'en ay point d'autre connoissance.] ^ ' 

Le Clerc reprinted the Hieronymian commentary in the twelfth 
volume of the Antwerp (Benedictine) edition of St Augustine's 
works in 1703^ but without any attempt, so far as tests have been 
able to show, to improve the text. His standpoint with regard to 
it was probably that of Garnier. 

In the fifth volume of the Martianay edition, published in 1706*, 
there is a note pointing out that the preface beginning Litteris tuis 
cursimis not to be found 'in veteri codiceVand thatthe commentary, 
though attributed in MSS to Jerome, is really to be ascribed to a 

1 Take such passages as 2 Cor. xi 9, 10; 1 Tim. iiii 8; 2 Tim. ii 24, and com- 
pare them with the text in my second volume. 

2 Memoires pour servir a VHistoire EccUsiastique des six Premiers Siecles, 
t. xn (Paris, 1707) pp. 115 f. He cites Lab. scr. t. i pp. 441, 794. In the latter 
passage L. quotes Aubertinus, p. 596, with approval. 

3 pp. 315—458. 

^ Sancti Eusehii Hieronymi Stridonensis Presbyteri Operum Tomus Quintus.., 
Parisiis, 1706, pp. 925—6. 

5 The reference would seem to be to Paris B.N. 1853 (saec. viri — ix), my M: 
the other old Paris MS, B.N. 9525 (saec. viii ex.), my E, was not in Paris till a 
century after this time. 


Pelagian, or to Pelagius himself. The editor had also noticed 
a reference to a copy of Pelagius on XIII epistles of St Paul in 
the catalogue of the library of St Riquier\ unobserved by any 
previous scholar. What follows must be quoted in Martianay's own 
words: ' Pt'opferea in quibusdcwi vetei^ibus codicihus praefixa est his 
Conimentariis concordia Epistolarum Beati Pauli, quam omnes 
veteres codices Pelagio trihuunt! The reference here is, I think, to 
the 'concordia' which occurs in Paris MS B.N. 1853 (saec. viii — ix), 
but neither in that MS nor in any other known to me, is it 
attributed to Pelagius. The use of our commentary by Primasius 
and Sedulius is then referred to, and he ends his note by explaining 
that when the name of Pelagius became unpopular, all the Pelagians 
in Gaul read the commentary under the name of Jerome. 

The erudite Jo. Albert Fabricius in his great Bihliotheca Graeca 
lib. iv (Hamburgi, 1708) p. 204, quoting Cotelier and others, gives 
' Pelagius sive potius Pelagianus quidam' as the author. 

Ceillier contented himself with a very brief resume of Garnier's 
researches in the following words: 'On croit avec beaucoup de 
vraisemblance que ce commentaire est celui-la meme que nous 
avons parmi les oeuvres de saint Jerome, puisqu'on y trouve la 
plupart des endroits qu'en ont cit6s saint Augustin et Marius 
Mercator, et qu'ils sont remplis d'erreurs pelagiennes. II est vrai 
qu'un des principaux passages cites par saint Augustin ne s'y voit 
plus: mais il est aise ou que P61age I'ait supprime lui-meme, ou 
qu'il en ait ete ote par Cassiodore qui, croyant que le pape Gelase 
etait auteur de ce commentaire, en avait purge I'Epitre aux 
Remains avec tout le soin possible, afin que d'autres corrigeassent 
a son exemple ce qu'il y avait d'errone dans ce Commentaire sur 
les autres Epitres de saint Paul^.' 

Yallarsi, at the end of his edition of the works of St Jerome, 
prefixes an 'admonitio' to his careless reprint of our commentaries, 
based almost entirely on Garnier's work and accepting its con- 

^ The exact words are: 'In codice Bibliothecae S. Richarii recensentur Com- 
mentarii Pelagii in XIII Epistolas Pauli.^ Zimmer learned this reference from 
Becker (see Pelagius in Irland p. 158) : the mention of it in Martiauay has es- 
caped notice. On the library of S. Eiquier, cf. L. Traube in Ahh. d. bayer. Akad. 
XIX (1892) pp. 326, 329, 331. 

^ Histoire Generale des Auteurs Sacres et Ecclesiastiques... noav. ed. t. vii (Paris, 
1861) p. 543 (original date, about 1750). 


elusions. He adds, however, two new facts on his own account, one 
when he points out that in the thirteenth century the commentary 
was known to a certain John of Verona as the work of Pelagius. 
The identity of this John is uncertain, Panvinius being of opinion 
that he is a certain John a Deacon, Pastrengicus that he was a 
'Presbyter Mansionarius' of the same name. It appears that he 
wrote a history, unprinted in Vallarsi's time, and perhaps still so, 
in which these words occur: ' Vidi ego ipse Joannes Commentarium 
Pelagii praedicti, super Epistolas Paidi, in quo licet midta bene et 
eleganter exponat, tamen suhtiliter infundit uenenum haeresis suae.' 
The second new fact is contained in the statement that the better 
manuscripts of the Hieronymian commentary are without the 
preface (to Heliodorus)*, As we shall see later, this is quite true. 

Jean-Baptiste Morel, priest of Auxerre, one of the acutest and 
most learned patristic scholars that ever lived, did not edit the 
commentary, but in his Elements de Critique, issued in 1766^ he 
proves that he had read it with great care: for he furnishes a 
number of emendations of the text, most of which are absolutely 
correct ^ but have been taken no notice of either by Vallarsi in his 
later edition, or by Migne in his reprint, or in fact by anybody at 

C. T. G. Schoenemann in his admirable Bibliotheca Historico- 
Literaria Patrum Latinorum gives an account of the editions of 
the Hieronymian commentary*. Towards the end of his account of 
Erasmus's edition, he makes a very pertinent suggestion, which, 
however, fell on deaf ears: 'Interim. . .optandum duxerim, ne doctum 
aliquem virum in hac sacrarum literarum parte hahitanteni pigeat 
denuo eadem excutere et imprimis inuestigare, annon fortasse assu- 
menta aliquot vilioris panni insint, quibas ineptus aliquis glossator 
vel compilator ad totius operis contemtum Erasmum provocaverit'^.' 
How much truth lies behind this suspicion will be apparent from 
a later part of the present volume*'. 

1 This fact had, however, been already observed by Martianay : see above, 
p. 21. 

2 More accessible in Migne's reprint. Premiere Encyclopedie Theologique, t. xlvh 
(Paris, 1866) pp. 969—1116. 

3 For example, in 1 Thess. iiii 13, Jieri ior fieri (Migne reprint, p. 1025). 
•» Vol. II (Lips. 1794) cap. 5 sect. 7 pp. 436 ff. 

5 p. 439. « pp. 35 f. etc. 


If editors and historians of Latin literature have given but 
scanty attention to the Hieronymian commentary, commentators 
on St Paul have made abundant use of it. It was a primary 
authority for a number of commentaries fi'om the sixth century 
onwards, and modern expositors have not neglected it. Among 
those of the nineteenth centur}^, Bornemann, commentator on the 
Epistles to the Thessalonians, Bp Lightfoot^ and Sanday and Head- 
lam on the Epistle to the Romans may be mentioned. Their attitude 
to the commentary is naturally that of Garnier. 

Garnier's view was disputed in a long paper by Klasen in the 
Theologische Quartalschrift for 1885 -. He compares the notes 
on Rom. v, as they appear in Augustine, Mercator, and the 
Hieronymian commentary, and acutely observes that, though 
there are many differences in form, the contents exactly harmonize^. 
He subjects the theology of the commentary to a thorough 
examination, and defends at great length the thesis that the work 
as we have it is all by one hand, and that hand is not Pelagius, but 
a Pelagian of the latest period, when Pelagianism was becoming 
Semi-Pelagianism. As most of the argument is theological, and 
the present writer is not a theologian, he cannot form an estimate 
of its value. Nor does he know any examination of the paper by 
a competent theologian, and he can only conclude either that the 
paper has been overlooked, or that the theologians regard it as 
negligible. Loofs in the earlier editions of his Leitfaden zura Studium 
der Dogmengeschichte is not sure whether the Hieronymian com- 
mentary has been worked over, or not, and avoids using it as an 
authority for Pelagius' theological views*. Gregory in his Textkritik 
des Neaen Testamentes^ holds that the Hieronymian commentary 
is 'sehr mutilirt.' 

It will be the lasting merit of the late Professor Heinrich Zimmer 
of Berlin that he drew the whole subject out of the state of 

1 Cf. especially bis commentary on the Epistle to the Galatians (Cambr. and 
London, 1865), pp. 222 f. 

2 Vol. Lxvii, pp. 244—317, 531—577. 

3 pp. 267—270. 

* In his article on 'Pelagius,' however, in the Herzog-Hauck Realencykl. (1904), 
and in the latest edition of the Leitfaden (1906), considerable use is made of it. 

* Leipz. 1909 (this part 1902) p. 810 : supplemented by later bibliography on 
p. 1353. 


stagnation into which it had fallen. His proper field was Keltic 
studies, and the book Pelagius in Irland, published in 1901 ^ 
was what he might have called a 'Seitenstiick.' He was a man of 
virile and suggestive mind, interested in problems both of history 
and philology. In the course of philological study, he came in 
contact with certain documents which shed light on the history of 
the Pelagius commentary, and with characteristic enthusiasm he 
turned to this subject, and worked it out in his own way. If it has 
fallen to me to correct some errors of his, and to introduce new 
elements into the problem, I wish nevertheless to record here with 
the utmost gratitude the stimulus I have received from his work. 
In fact, when I took up the subject, it was with the intention simply 
of working out certain lines of investigation which he had suggested. 
Nor is it only to myself that stimulus has come from his work. 
A number of scholars have been led to interest themselves in the 
problem, such as Turner, Loofs'^ Riggenbach, Hellmann, Morin 
and De Bruyne. 

Zimmer* points out that the Book of Armagh, the well known 
MS in the library of Trinity College, Dublin, written in 808, con- 
tains a prologue to the Epistles of St Paul, a prologue to the Epistle 
to the Romans, and separate arguments for each of the epistles, 
nearly all of which are specificall}^ assigned to Pelagius'*. He draws 
attention also to a MS of the early part of the eighth century in 
WUrzburg^. This is a MS of the Epistles of Paul in Latin, which 
is heavily glossed. Most of the glosses, actually 949®, are stamped 
pi, and one of them coincides with a comment quoted by Marius 

^ Berlin, Weidmann, pp. viii + 450. 

2 See his article in Herzog-Hauck's Realencykl? xv (Leipz. 1904) pp. 747 — 
774; also the supplementary and corrective article in Bd. xxiv (Leipz. 1913) 
pp. 310—312. 

2 In this and following paragraphs I follow very closely the wording of my 
lecture, published in the Proceedings of the British Academy vol. ii (1905 — 1906) 
pp. 409 — 439, which in its separate form has been for some time out of print. 

■* But see page 17, above, for the proof that Ussher had noticed this fact much 
earlier, as also S. Berger, Histoire de la Vulgate (Paris, 1893) pp. 32 f.; Les Pre- 
faces jointes aux livres de la Bible (Paris, 1902) p. 26. 

5 Since published in photographic facsimile by L. C. Stern (Halle a. S. 1910). 
The date I take from W. M. Lindsay, Notae Latinae (Cambr. 1915) p. 493, who 
says : 'glosses most of them patently much later than the text.' 

* Zimmer, p. 112, for statistics. 


Mercator^ The majority of these, actually 8-iO, are to be found in 
our Pseudo-Jerome commentary, but the remainder are not to be 
found there. There are also, however, 348 glosses, with no author's 
name attached, that are to be found in the Pseudo-Jerome. The 
total reaches altogether 1311^. Zimmer has published a list of the 
glosses that are connected with the Pelagian commentary, and 
has added convenient symbols, first, to show when the gloss, though 
denominated pi, is not to be found in our Pseudo-Jerome, and 
second, when, though anonymous, it is to be found there. He has 
also discovered on examination that some of the glosses, labelled pi, 
are not really by Pelagius, but are derived fi*om the commentary 
which was first printed in 1537 as a work of Primasiusl The 
original compiler of these glosses appears to have possessed an un- 
mutilated Pelagius, since many passages which we cannot find in 
Pseudo-Jerome, are to be found either in Pseudo-Primasius or in 
Sedulius Scottus, both of whom made extensive use of Pelagius*. 
Some of those passages are to be found in both of these compilers. 
From his study of the anonymous glosses in this MS, Zimmer 
found that the so-called Primasius commentary was sometimes 
used by the compiler, though not cited by any name. This com- 
mentary Haussleiter had proved to have no connexion with 
Primasius^ and Zimmer claimed to be the first to point out that 
the principal source used in it is the Pelagius commentary, in an 
unmutilated state^; but here he overlooked the work of some 
seventeenth century scholars'', though the Primasian authorship 
was then hardly doubted ^ Zimmer, however, deserves all credit for 

1 On Rom. v 15 ; Zimmer, p. 40 : see the next chapter. 

2 Zimmer, p. 132. » pp. 4-5, 68, 129. 

■* Aubertin ap. Labbe, 1. 1 p. 796, and Simon, pp. 336 ff., 380, had pointed this out 
long before. Before Simon the Benedictines of St Maur had observed it, as regards 
Primasius; seefheix Augustine, t. x (Paris, 1690) pruef. : Primasium ex hoc commen- 
tario (i.e. Ps.-Hier.) non pauca desumsisse, fonte interim, unde ilia duceret, non 
indicate, ah eniditis obseruatum est. See also H. B. Swete, Theodori Episcopi 
Mopsuesteni in epist. B. Pauli Commentarii vol. i (Cambr. 1880) p. xlv n. 1. 

* Zahn's Forschungen z. Gesch. des iieutestamentl. Kanons iv (Erl. & Leipz. 1891) 
pp. 24—35. 

8 His words are 'was bisher noch nirgends erkannt wurde' (p. 122). 

7 See the above note. 

^ Yet it was noticed that the conim. on Hebr. was identical with that of Haymo 
(Migne, P.L. cxvii) and Remigius (not in Migne): see Tho. Gataker, Adv. Misc. n 20 
('si is est'); Simon, p. 368; Westcott's ed. of Hebrews (Lond. 1889) p. vii. 


the detailed manner in which he has dealt with the question. He 
shows, for example, that the borrowing is much easier to detect in 
the later epistles than in the Epistle to the Romans, and that the 
whole commentary^ is an anonymous anti-Pelagian revision of the 
Pelagius commentary 2. He also allows that in the Wiirzburg codex 
a portion of the anonymous glosses, which ultimately belongs to 
Pelagius, may have come to it through pseudo-Primasius, shows 
that sometimes there are errors in the citations^, and, further, proves 
that the compiler of the glosses sometimes abridged his sources*. 

As to the date and place of origin of the pseudo-Primasius 
commentary, Zimmer argues that Pelagianism on the Continent 
ended with the decrees of the Synods of Orange and Valence in 
529, and that this date provides a terminus ante quern for a commen- 
tary containing polemics against the Pelagians. He seeks further 
to draw from the allusion 'fuerunt Hunni usque ad Attilam^' an 
argument that the compiler could not have lived long after the 
death of Attila (454), and that the last third of the fifth century is 
the latest possible date for the commentary. He thinks it came 
into being in South Gaul or North Italy in connexion with the 
semi-Pelagian controversies, and proposes to identify it with the 
commentary which Cassiodorus knew as attributed by some to 
Pope Gelasius and which he himself afterwards revised. This 
reasoning is able and learned, but fallacious, as we shall see. There 
is, however, no need to disagree with Zimmer as to the date when 
the pseudo-Primasius commentary came to Ireland, namely in 641''. 
He further mentions that this commentary is used in the Haymo- 
Remigius commentary of the ninth century^. 

Besides the Wiirzburg codex, Zimmer drew attention to two 
other MSS of the Epistles in Latin, which provide similar material, 
though much inferior in bulk. They are Vienna MS 1247, written 

1 Excepting, of course, Hebr. - pp. 121 ff. 

3 pp. 127 ff., 133. * p. 133. 

5 Migne, P.L. lxviii, i41 b; Zimmer, p. 135. 

« Zimmer, p. 137. 

7 pp. 135, 162. It is much to be desired that the extent of this indebtedness 
should be made the subject of a special monograph. In the course of the slight 
study I have been able to give to the question, I have not noticed that the indebted- 
ness extends far. Also Haymo worked up his sources, and did not transfer them 


in 1079, and Berne MS A 73, of the thirteenth century. The former 
contains 203 direct citations from Pelagius: of these 167 are to be 
found in pseudo-Jerome, while 36 are there lacking. In 57 places 
also where no name is attached, words are given which appear in 
our pseudo-Jerome. Both MSS give the same prologue to the 
Epistles as is given by the Book of Armagh, and under Pelagius's 
name. The total number of independent glosses in the Wiirzburg 
and Vienna MSS together reaches 1535'. Zimmer also refers to 
citations of Pelagius in the Irish Canons of the end of the seventh 
or the beginning of the eighth century-. 

He next points out that there is evidence for the existence in 
the ninth and tenth centuries of MSS of Pelagius's commentary in 
three ^ libraries, all of them connected with the Irish mission to the 
Continent, at St Riquier, Lorsch, and St Gall^ Of these the St Gall 
MS appears to have survived in the anonymous 73, which will be 
further described immediately. 

He then shows quite clearly, in opposition to the ruling view 
of Gamier and Simon, that the pseudo-Jerome commentary shows 
no prevailing anti-Pelagian tendency^, and that it cannot therefore 
be the revision by Cassiodorus. In this respect pseudo-Jerome is 
to be contrasted with pseudo-Primasius, Sedulius, the Wiirzburg 
and Vienna glosses, to mention the other commentaries which 
employ Pelagius throughout. He proceeds to show by copious 
illustrations that frequentl}^ all other authorities have preserved 
the true text against pseudo-Jerome, and suggests — quite rightly, 
as will afterwards be proved — that many of the errors of that form 
will vanish when MSS of it are collated". One of the best parts 
of Zimmer's book is his proof that Pelagius wrote no commentary 
on the Epistle to the Hebrews, though he regarded it as Pauline''. 
Fresh arguments will be adduced to confirm his conclusion^ 

1 Zimmer, p. 155. 2 p, 152, etc. 

^ Prof. S. Hellmann adds a fourth, that at Murbach: the entry in the catalogue 
(dated about 840) (ed. Bloch, p. 271) is: — 210, Exposicio Pelagii in epistolas Pauli. 

* Zimmer, pp. 156 fif. » pp. 164 ff. 
^ Zimmer, pp. 169 — 175. 

^ pp. 178 ff. In this connexion he gives Haussleiter the credit of observing that 
the commentary on Hebrews in ps.-Primasius is identical with that in Haymo- 
Remigius ; but this had been observed long before : see p. 26 n. 8 above. 

* ch. vi. Cf. also De Bruyne in Revue Biblique, nouv. ser. xii (1915) p. 372. 


Zimmer identifies the pseudo- Jerome commentary with that 
mentioned by Cassiodorus as containing short notes which were 
ascribed by some to Jerome. The second commentary mentioned 
by Cassiodorus he identifies with ' Ambrosiaster.' The first of the 
three anonymous commentaries mentioned by Cassiodorus, which 
Gamier and others have, in my opinion rightly, believed to be the 
unmutilated commentary of Pelagius, Zimmer considers to have 
been identical with our pseudo-Primasius. With regard to the re- 
vision by Cassiodorus and his pupils, he is of opinion that it has 

His view as to the manner of production of the pseudo- Jerome 
commentary must be mentioned. He considers that some one in 
the first half of the fifth century, before the suppression of 
Pelagianism, wrote out notes from the Pelagius connnentary in a 
copy of the Epistles of Paul in Latin, that these notes fell into the 
hands of a man who knew nothing of Pelagius and was no heretic 
hunter, about the end of the fifth or the beginning of the sixth 
century, and that this man ascribed them to Jerome and edited 
them as his'. 

Zimmer's researches'^ of which we have here given a rather in- 
complete account, were crowned by the discovery of what appears 
to be the St Gall MS above referred to. Manuscript no. 73 in the 
Stifts-Bibliothek contains an anonymous commentary on fourteen 
Epistles of St PauP. Its special characteristics are illustrated in 
great detail, by its discoverer, who compares it with pseudo- Jerome. 
It has some distinct differences from the published form attributed 
to Jerome. For example, it gives a long quotation in its proper 
place, which was already known from Augustine and Mercator^ 
but is wanting in pseudo-Jerome, and in text it agrees with 
Augustine against Mercator. In the commentary on First Corin- 
thians it frequently lacks one of the two explanations which 
pseudo-Jerome offers for one verse, and this Zimmer attributes to 
the editing of some Irish scholar. But the St Gall MS is not a pure 

1 pp. 200 ff. 

- Some of them were anticipated in his article, ' Keltische Kirche in Britannien 
und Irland' in Bd. x (1901) (especially p. 211) of Herzog-Hauck's Realenojhlo- 
pddie : the article was afterwards published as a book in an English translation, 
Celtic Church in Britain and Ireland, by A. Meyer (London, 1902). 

3 Zimmer, pp. 219—279. * See below, pp. 36 f., 45 f. 


Pelagiiis. It is heavily interpolated from known sources, especially 
in the commentaries on Ephesians, Titus, and Philemon, where 
passages from the genuine commentaries of Jerome on these epistles 
are interwoven with the Pelagian original, without acknowledgment. 
There are also citations from Augustine and Gregory the Great. 
These additions Zimmer regards not as specialities of the Irish 
recension of Pelagius, but as peculiar to the St Gall MS or its 
originals The Irish appear from early times to have added notes 
to their copies of Pelagius, and thus in Irish circles passages came 
to rank as Pelagius which were not really by him. Zimmer suggests 
that Sedulius^ and the Wurzburg and Vienna MSS^ are thus inter- 
polated. After sketching the only plan by which an edition of 
Pelagius could then be made, Zimmer presents his readers with 
a collation of the St Gall MS made with the text of pseudo- 
Jerome ^ The collation is fairly exact as far as it goes, but is 
defective to an incredible degree'. In spite of the care he exercised ^ 
Zimmer appears to have been physically incapable of making a 
really exact and full collation of a manuscript. Yet by this publi- 
cation he did a great service to the study both of pseudo-Jerome 
and of Pelagius, as he at once removed a large number of the cor- 
ruptions of the printed text of the former. 

The next step in investigation was taken by Dr C. H. Turner 
in a most valuable review in the Journal of Theological Studies''. 
He joined issue with Zimmer on one point only, namely his view 
as to the pseudo-Primasius commentary. This commentary, as we 
have seen, Zimmer seeks to identify with the first commentary 
mentioned by Cassiodorus, from which Cassiodorus says he after- 
wards removed the Pelagian characteristics. Turner rightly objects 
that the pseudo-Primasius is 'definitely and consciously anti- 
Pelagian throughout,' and that it is in the commentary on Romans 
that Pelagius has been specially re-written. He proceeds: 'Pseudo- 
Primasius is based on Pelagius : if Garnier and the scholars who 

* My own researches confirm this conclusion ; see below, pp. 239 ff. 
- I should not agree here, except perhaps in a very limited sense. 

3 Zimmer, pp. 268—271. 

■* Except that Eom. i and v 12 — 21 and Hebr. are copied in extenso. 

* Hardly less than 2000 omissions have been noted. 

* He tells us that he revised the collation with the original (pp. 449 f.). 
7 Vol. IV (1902—1903) pp. 1.32—141. 


have followed him are right, Cassiodorus revised Pelagius. Pseudo- 
Primasiiis is an anti-Pelagian edition of Pelagius : so was that of 
Cassiodorus. Pseudo-Primasius has revised his original more on 
the doctrinal than on the linguistic side, more in the Epistle to 
the Romans than in the other epistles-; Cassiodorus, "in order to 
remove far the error of heresy," purged the Epistle to the Romans 
with all the curiositas that he could, leaving the rest of the revision 
to his pupils, whose work will doubtless have been much more 
perfunctory than their master's. Pseudo-Primasius adds to the 
genuine Pelagius on the thirteen epistles a commentary on the 
Hebrews, which depends on Chrysostom's Homilies^ : but it was 
Cassiodorus who, in order to provide a commentary on an epistle 
which both Ambrosiaster and Pelagius had neglected, caused a 
certain Mutianus to translate these Homilies of Chrysostom into 
Latin^ The correspondence appears to be exact: pseudo-Primasius 
is surely nothing else than the new and standard commentary on the 
completed Pauline epistles evolved out of Pelagius and Chrysostom 
by Cassiodorus and his monks of Vivarium''.' 

Though Zimmer had very fully reviewed the mediaeval authori- 
ties which made use of Pelagius, there was one at least which 
escaped him. Zmaragdus, abbot of St Mihiel at the end of the 
eighth century and the beginning of the ninth, had compiled out 
of the writings of some twenty authors, whom he names in his 
preface, a commentary on the lessons read in church, and indicated 
his borrowings by symbols in the margins. This commentary was 
published at Strasbourg in 1536, and reprinted in Migne's Latin 
Patrology, vol. cii, in 1851. In this reprint Dom Pitra had called 
attention to the fact that Pelagius was one of the authors used and 
named by Zmaragdus, and Primasius another, and had brought 
into clear relief some instances of disagreement between his quota- 
tions and the pseudo-Jerome. Pitra's work had been overlooked by 
Zimmer, and the connexion of Zmaragdus with Pelagius was re- 
discovered about the same time by Riggenbach, Hellmann, Morin, 
and myself. 

1 Zimmer, p. 122. - Zimmer, pp. 183 — 195. ■* Zimmer, p. 202. 

•* The suggestion with regard to pseudo-Primasius onHebr.,here made by Turner, 
is mistaken. Pseudo-Primasius on Hebr. is really the work of Haymo of Auxerre, 
and quotes Cassiodorus by name. Its union, therefore, with the pseudo-Primasius 
on the other epistles cannot have begun before the ninth century. See below, pp. 321 £f. 


Riggenbach, in 1905, published an article pointing out the value 
of Zmaragdus in this respects He was able to identify the original 
editor of Zmaragdus as Caspar Hedio-. He very acutely observed, 
from Pitra's collation of the symbols in a Boulogne MS of Zmaragdus, 
that the symbol P in the manuscript was used by Zmaragdus only 
in commenting on the Epistles of Paul, and PR[ or PR only in 
comments on the Apocalypse, and that P therefore means Pelagius, 
PRI or PR Primasius. He rightly, concludes that Zmaragdus knew 
no Primasius on the Epistles, and that the first editor had resolved 
the symbol P wrongly, when he took it everywhere to mean 
Primasius. Riggenbach's paper, though brief, is also in other re- 
spects extremely suggestive. He shows that Zmaragdus and pseudo- 
Primasius agree on occasion in providing a text of Pelagius, which 
is different from, and obviously better than that provided by either 
pseudo- Jerome or the St Gall MS^ and he conjectures that the 
two last authorities represent a definite recension of the original 
commentary. He also draws up a very useful list of the quotations 
in Zmaragdus labelled P, as far as he could learn them from Pitra's 
reports. He has examined some attributions to ' Primas.' in the 
printed text, and has rejected them on internal evidence. On the 
whole he was disposed to regard pseudo- Jerome as an abridged 
form of the original Pelagius. He adds a new glossed MS to the 
list provided by Zimmer, namely Berlin Codex Phillippicus 1650 
(saec. XI ex., formerly of St Vincent at Metz), which introduces some 
glosses with the letters ' Pel': this MS I also had noted. 

Hellmann in his Sedulius Scottus, published early in 1906, took 
occasion to include a careful study of the use of Pelagius made by 
Sedulius in his own commentary. The value of his work was 
heightened by the fact that he examined the MSS of Sedulius's 
work as well as the printed text. He overthrows Zimmer's conten- 

^ ' Unbeachtet gebliebene Fragmeute des Pelagius-Kommentars zu den Paulini- 
schen Briefen ' (Beitriige zur Farderung christlicher Theologie ix 1. Giitersloh). I owe 
my first knowledge of this paper to the late Rev. Prof. John E. B. Mayor. 

2 One of the lesser Reformers, and Miinsterprediger at Strasbourg: see Corpus 
Reformatorum xciv {1911) p. 213 n.; P. Kalkoff, W. Capita im Dienste Erzbischof 
Alhreehts von Mainz (Berlin, 1907) pp. 38 f. etc. 

^ Of course, as reported by Zimmer, from whom ex silentio inferences are never 
safe. Pseudo-Jerome as printed, also, is a very different thing from pseudo-Jerome 
as textually restored. 


tion that the St Gall MS with Sedulius and the Wiirzburg and 
Vienna MSS represents the Irish tradition of Pelagius, as against 
the Continental tradition represented by pseudo-Jerome and 
pseudo-Primasius, and shows on the contrary that there is a real 
relationship between the St Gall MS and pseudo-Jerome over 
against all other authorities for Pelagius^ This relationship shows 
itself in community of corruption, in cases where the true text 
can be elicited from pseudo-Primasius, Zmaragdus, and Sedulius 
Scottus. Hellmann handles much material of varying quality with 
great skill, and makes many suggestive remarks on this difficult 
subject. He also shows that Isidore of Seville used Pelagius in one 
form or another^ that several glossed MSS of St Paul's Epistles 
contain Pelagian matter (Clm. 9545 [saec. x]^ 18530 [saec. XI — xii], 
Berlin, theol. fol. 481, Einsiedeln 16, Karlsruhe Augiensis Lxxxiii 
[saec.xi]^), and that the so-called Pelagian prologues occur — though 
without his name — in the oldest Vulgate MSS we possess. 

I think I have now given some account of every important 
publication on this subject prior to the appearance of my own 
earliest article on it (July, 1906). Since about 1904 I had been 
taking an active interest in the matter, had collated portions of the 
Bodleian MS of Zmaragdus and certain Pelagian prologues, and 
had also in 1905 and 1906 put together a tentative list of MSS 
with some bearing on the problem of the Pelagian commentary ^ 
As research proceeded I published a number of articles on special 
points, as well as two summaries of progress in the Fi^oceedings of 
the British Academy, volumes ii (1907) and vii (1916). If I do not 
at this point chronicle the valuable articles published by other 
scholars since the summer of 1906, I hope my attitude will not be 
misunderstood. As these articles have in great part grown out of 
my own, it seems more suitable to the scheme of the present work 
to incorporate their results at the proper points in the succeeding 

1 Here again the qualifications stated in the above note, p. 32 n. 3, must be kept 
in mind. 

2 pp. 152, 184. 

3 pp. 186—190. 

■* p. XV. 

5 Published in Journal of Theological Studies vol. vii (1905—1906) pp. 568—575. 
S. P. 3 



From the previous chapter it will have been gathered that the 
Pelagius commentary encountered some strange experiences in the 
course of its history. To identify it among existing commentaries 
is a work requiring caution. The method adopted here is to con- 
front the claimants in succession with the quotations made by 
Augustine and Mercator, these being contemporary or almost con- 
temporary witnesses to its text. Augustine indeed became ac- 
quainted with it at least as early as 412, some three years after 
its completion. The Mercator quotations have a value all their 
own, though they show some verbal differences from those of 

The claimants are four in number, Pseudo-Jerome ; the text in 
St Gall MS 73 ; the text in Paris MS 653 ; and the text in Karlsruhe, 
cod. Augiensis, cxix, and Balliol College, Oxford, MS 157. These 
authorities will be described in full later. It will be sufficient at 
this stage to note that Pseudo-Jerome is, as here printed, critically 
reconstructed from a number of MSS, and not merely copied from 
a printed edition ; that the text of St Gall MS 73 is given ac- 
cording to Zimmer's collation, revised by me with the original ; 
that the Paris MS 653 is an anonymous MS, written in the Veronese 
district late in the eighth century, containing a long commentary 
on fourteen Epistles of St Paul; that the Karlsruhe MS was written 
at Reichenau about the same date and contains a short anony- 
mous commentary on thirteen Epistles of St Paul, while the Balliol 
MS, an Italian product of the mid-fifteenth century, though it bears 
the name of Jerome in its title, contains substantially the same 
commentary as the Karlsruhe MS. 

CH. Il] 




{De Pecc. Mer. et Rem. in ii 2 to 
'perierunt': iii 5 to 'alieua.' (C.S.E.L. 
LX pp. 129, 132) : cf. viii 15, 16 (p. 141), 
X 18 (p. 144), De Pecc. Orig. xxi 24 
(C.S.E.L. xxxxii p. 183), Epist. 190, 22 
(C.S.E.L. LVii p. 158), Op. imperf. c. 
lulian. I 56, c. lul. Pelag. vi vii 18) 

hi autem qui contra traducem peccati 
sunt, ita illam impugnare nituntur: 
'si Adae,' inquiunt, 'peccatum etiam 
non peccantibus nocuit,ergo etChristi 
iustitia etiam non credentibus prod- 
est, quia similiter, immo et magis, 
dicit per unum saluari, quam per 
unum ante perierunt.' deinde aiunt : 
'si baptismus mundat antiquum illud 
delictum, qui de duobus baptizatis 
nati fuerint debent hoc carere pecca- 
to; non enim potuerunt ad posteros 
transmittere quod ipsi minime habue- 
runt. illud quoque accedit {most MSS 
accidit), quia, si anima non est ex 
traduce, sed sola caro, ipsa tantum 
habet traducem peccati, et ipsa sola 
poenam meretur.' iniustum esse di- 
centes, ut hodie nata anima non ex 
massa Adae tam antiquum peccatum 
portet alienum, dicunt etiam nulla 
ratione concedi, ut deus, qui propria 
peccata remittit, imputet aliena. 

remittit] dimittit Epist. 190. 

imp. aliena] unum inp. alienum 
Epist. 190. 


in Rom. v 15 

hie manifeste docet quia non gene- 
raliter de omni homine dicit, dicens : 
unius delicto multi mortui sunt, 
quia communi et naturali morte non 
solum peccantes, sed et iusti mori- 

{De Gestis Pelagii xvi 39 (C.S.E.L. 
XXXXII p. 94)) (paraphrastic) 

hoc quod scriptum est [Rom. viiii 16], 
non ex persona Pauli adserit dictum, 
sed eum uoce interrogantis et redar- 
guentis usum fuisse, cum hoc diceret, 
tamquam hoc dici utique non de- 

in Rom. viiii 16 

unde intellegitur quia 
gautis uoce utitur et 
potius quam negantis. 

hie interro- 





{De Pecc. Mer. iii xii 21 (C.S.E.L. lx 
p. 148)) (paraphrastic) 

exempla iam praecesserant et uirorum 
quos uxores et feminarum quas mariti 
lucri fecerant Christo et paruulorum 
ad quos faciendos Christianos uolun- 
tas Christiana etiam unius parentis 

in 1 Cor. vii 14 

exemphim refert quia saepe conti- 
gerit ut lucri fieret uir per mulierem 
.... saepe enim sic contigerat ut filii 
ilium parentem qui crediderat seque- 

While from the second and third (paraphrastic) quotations one 
might very well conclude that Pseudo-Jerome was the original 
Pelagius, the first long and exact quotation clearly proves that 
Pseudo-Jerome cannot be the original Pelagius, for here the two 
authorities differ entirely. Let us next compare Augustine with 
St Gall MS 73. 


hi autem qui contra traducem peccati 
sunt, ita illam impugnare nituntur : 
'si Adae,' inquiunt, 'peccatimi etiam 
non peccantibus nocuit, ergo et Christi 
iustitia etiam non credentibus prod- 
est, quia similiter, immo et magis, 
dicit per unum saluari, quam per unum 
ante perierunt.' deinde aiunt: 'si 
baptismus mundat antiquum illud 
delictum, qui de duobus baptizatis 
nati fuerint debent hoc carere peccato ; 
non enim potuerunt ad posteros trans- 
mi ttere quod ipsi minime habuerunt. 
illud quoque accedit {inost MSS acci- 
dit), quia, si anima non est ex traduce, 
sed sola caro, ipsa tantum habet tra- 
ducem peccati, et ipsa sola poenam 
meretur. iniustum esse dicentes, ut 
hodie nata anima non ex massa Adae 
tam antiquum peccatum portet 
alieuum, dicunt etiam nulla ratione 

St Gall MS 73 

Plus praeualuit iustitia uiuificando 
quam peccatum in occidendo, quia 
Adam tantum se et suos posteros 
interfecit, Christus autem et qui erant 
tunc in corpore et posteros liberauit. 
hi autem qui contra traducem peccati 
sunt, ita ilium inpugnari nituntur : 
'si Adae,' inquiunt, 'peccatum etiam 
non peccantibus nocuit, ergo et Christi 
iustitia etiam non credentibus prod- 
est, quia similiter, immo et magis, 
dicit per unum saluari quam per unum 
ante perierunt.' delude aiunt: 'si 
baptismum mundat antiquum illud 
delictum, qui de duobus baptizatis 
nati fuerint debent hoc carere peccato ; 
non enim potuerunt ad filios trans- 
mittere quod ipsi minime habuerunt. 
illud quoque accidit, 

quia, si anima non est ex traduce, 
sed sola caro, ipsa tantum habet tra- 
ducem peccati, et ipsa sola poenam 
meretur. iniustum esse dicentes, ut 
hodie nata anima non ex massa Adae 
tam antiquum peccatum i^ortet quam 
alienum, dicunt etiam nulla ratione 




concedi, ut deus, qui propria peccata 
remittit, imputet aliena. 

(as above, p. 35) 

concedi, ut deus, qui propria peccata 
remittit, inputet aliena. hie manifesto 
docet quia non generaHter de omni 
homine dicit dicens : uniusdelicto 
multi mortui sunt, quia communi 
et naturali morti non solum peccantes 
sed et iusti moriuntur. 

in Rom. viiii 16 
(no difference from Ps.-Hier. p. 35) 

exempla iam praecesserant et uirorum 
quos uxores et feminarum quas mariti 
lucri fecerant Christo et paruulorum 
ad quos faciendos Christianos uoluntas 
Christiana etiam unius parentis eui- 

in 1 Cor. vii 14 
exemplum refert quia saepe contigerit 
ut lucri fieret {corr. lucrificaretur) uir 

per mulierem saepe enim si {corr. 

sic) contigerat ut filii ilium parentem 
qui crediderat sequerentur. 

The case here is different from the last. While it is true that 
there are some slight textual differences between the two columns 
of text, these are easily explicable : ilium for illam is a palaeo- 
graphical error ; inpugnari for inpugnar'e is due to carelessness ; 
baptismus would seem to be a correction by Augustine or his scribes 
of the less pure form baptismum, which Pelagius doubtless wrote ^ ; 
posteros and the insertion of quam are conscious alterations on 
the part of revisers. But when all these facts are admitted, the 
St Gall MS still contains the whole passage quoted by Augustine. 
It also contains the passage which Pseudo-Jerome substitutes or 
seems to substitute for the original Pelagius, but at this stage, 
whatever our suspicions may be, we cannot, on the evidence sub- 
mitted, deny that the St Gall MS may represent the original 

Augustine Paris MS 653 

Hie manifeste docet quia non genera- 
liter de omne {sic) homine dicit, di- 
cens: unius delicto multi mortui 
sunt, quia communi et naturali 
morte non solum peccantes, sed et 

1 See below Chap, ill (e) p. 95. 




Hi autem, qui contra tra- 
ducem peccati sunt, ita illam impug- 
nare nituntur: 'si Adae,' inquiunt, 
'peccatum etiam non peccantibus 
nocuit, ergo et Christi iustitia etiam 
non credentibus prodest, quia simil- 
iter, immo et magis, dicit per unum 
saluari, quam per unum ante perie- 
runt.' deinde aiunt: 'si baptismus 
mundat antiquum illud delictum, qui 
de duobus baptizatis nati fuerint de- 
bent hoc carere peccato ; non enim 
potuerunt ad posteros transmittere 
quod ipsi minime habuerunt. illud 
quoque accedit {7nost MSS accidit), 
quia, si anima non est ex traduce, sed 
sola caro, ipsa tantum habet traducem 
peccati, et ipsa sola poenam meretur.' 
iniustum esse dicentes, ut hodie nata 
anima non ex massa Adae tam anti- 
quum peccatum portet alienum, di- 
cunt etiam nulla ratione concedi, ut 
deus, qui propria peccata remittit, 
imputet aliena. 

iusti moriuntur. [This portion comes 
after multi mortui sunt(670A3), 
and is in turn followed by the rest of 
Rom. V 15] then : — 
marginal note with reference to plures 
= quia plures inuenit quos seduceret : 
then Bwn. v 16 as far as donum 
followed by : — 

Plus praeualuit ' iustitia in uiuificando 
quam peccatum in occidendo, quia 
Adam tantum se et suos posteros 
interfecit, Christus autem et se et qui 
erant tunc (in) corpora et posteros 
liberauit. Hii autem, qui contra tra- 
ducem peccati sunt, ita illam inpug- 
nare nitimtur: 'si Adae,' inquiunt, 
'peccatum etiam non peccantibus 
nocuit, ergo et Christi iustitia etiam 
non credentibus prodest, quia simil- 
iter, immo et magis, dicit per unum 
saluari quam per unum ante perie- 
rant.' deinde aimit: 'si baptismum 
mundat antiquum illud delictum, qui 
de duobus baptizatis nati fuerint de- 
bent hoc carere peccato; non enim 
potuerunt ad filios transmittere 
quod ipsi minime habuerunt. illud 
quoque accidit, 

quia, si anima non est ex traduce, sed 
sola caro, ipsa tantum habet traducem 
peccati, et ipsa sola poenam meretur. 
iniustum esse dicentes, ut hodie nata 
anima non ex massa Adae tam anti- 
qua peccata portet aliena, asse- 
runt etiam nulla ratione concedi, ut 
deus, qui propria peccata dimittit, 
inputet aliena. 

Aliter: Adam solam formam fecit 
delicti, Christus uero et gratis peccata 
remisit et iusti tiae dedit exemplum 
uiuendi. Aliter: Ostendit maiorem 
uim esse gratiae quo (sic) in se homines 
trahat, quam peccati, id est, diabuli: 

Really p#ualuit. 


nam diabulus Aeuam decepit et per 
exemplum illius ad alios cucurrit ; 
gratia uero et multos inuenit quos 
credentes iustificauit, et permultos (w 
per multos) facile in se alios inui- 

(paraphi'astic) in Rom. viiii 16 

(see above, p. 35) (no difference from Ps.-Hier. p. 35) 

(paraphrastic) in 1 Cor. vii 14 

exempla iam praecesserant et uirorum exemplum refert quia saepe contigerit 

quos uxores et feminarum quas mariti ut lucri fieret uir per mulierem 

lucri fecerant Christo et paruulorum sic contigerat ut filii ilium parentem 

ad quos faciendos Christianos uolun- qui crediderant sequerentur. 
tas Christiana etiam unius parentis 

The situation here is not very unlike that in the last case. 
The long extract quoted by Augustine is found also in this MS, 
Paris 653. There are a few trifling differences in text; the frequent 
error hii for hi, perierant for perierunt, haptisrnum for baptismus, 
filios for posteros, antiqua peccata portet aliena for antiquum, pec- 
catum portet alienum, asserunt for dicunt and dimittit for remittit^. 
The situation in Rom. viiii 16 is identical with what we have 
found in the documents previously adduced. In 1 Cor. vii 14 it is 
also identical, except for the absence of saepe enim before sic con- 
tigerat, and the corruption crediderant for crediderat. All things 
considered, we cannot as yet refuse to this document the title to be 
called Pelagius, though we may well doubt the primary character 
of a form of the commentary that furnishes in all five notes on this 
passage, three of which are absent from the two authorities already 
examined. Moreover, one of these notes, namely that beginning 
Adam solam formam fecit, appears in Pseudo- Jerome at a somewhat 
later point in the commentary, where also Paris MS 653 has it a 
second time, namely after the last clause of Rom. v 16 (Migne 670 b). 
We shall now compare our quotations with the anonymous Reichenau 
MS cxix and the Balliol College MS 157 together. 

1 Compare the variant in Aug. Epist. 190 recorded above, p. 35. 





hi autem qui contra traducem pec- 
cati sunt, ita illam impugnare nitun- 
tur: 'si Adae,' inquiunt, 'peccatum 
etiam non peccantibus nocuit, ergo et 
Christi iustitia etiam non credentibus 
prodest,quia similiter, imrao et magis, 
dicit per unum saluari, quam per 
unum ante perierunt.' deinde aiunt : 
'si baptismus mundat antiquum illud 
delictum, qui de duobus baptizatis 
nati fuerint debent hoc carere pecca- 
to; non enim potuerunt ad posteros 
trausmittere quod ipsi minime habue- 
runt. illud quoque accedit {most MSS 
accidit), quia, si anima non est ex 
traduce, sed sola caro, ipsa tautum 
habet traducem peccati, et ipsa sola 
poenam meretur.' iniustum esse di- 
centes, ut hodie nata anima non ex 
massa Adae tarn antiquum peccatum 
portet alienum, dicunt etiam nulla 
ratione concedi, ut dens, qui propria 
peccata remittit, imputet 

Cod. Aug. cxix and 
Cod. Ball. 157 

Plus praeualuit iustitia in uiuifi- 
cando quam peccatum in occidendo, 
quia Adam tantum se et suos posteros 
interfecit, Christus autem et qui erant 
tunc in corpore et posteros liberauit. 
hi^ autem qui contra traducem pec- 
cati sunt, ita illam impugnare nitun- 
tur: 'si Adae,' inquiunt, 'peccatum 
etiam non peccantibus nocuit, ergo et 
Christi iustitia etiam non credentibus 
prodest, quia similiter, immo et magis, 
dicit 2 per unum saluari, quam^ per 
unum ante perierunt*.' deinde aiunt : 
' si baptismum mundat antiquum ill ut^ 
delictum, qui de duobus baptizatis 
nati fuerint debent hoc carere pecca- 
to ; non enim potuerunt ad filios 
transmittere^quod ipsi minime habue- 
runt. illut quoque accidit, 

quia, si" anima non est ex 
traduce, sed sola caro, ipsa tantum 
habet traducem peccati, et ipsa sola 
paenam* meretur.' iniustum esse di- 
centes, ut hodie nata anima non ex 
massa Adae tam antiquum peccatum 
portet alienum, dicunt etiam ^ nulla 
ratione concedi, ut deus qui propria 
homini^" peccata remittit imputet 


in Rom. viiii 16 
(no diflFerence from Ps.-Hier. p. 35, 
except that Ball. om. et redar- 
guentis) ^'^. 

1 hii Ball. - decet Ball. ^ qua Ball. * perierant Ball. 

* Such forms are a specialty of Aug. among my MSS, see p. 208. 
^ trasmittere Ball. "^ quia si] quasi Ball. * poenam Ball. 

9 etiam om. Ball. '" homini om. Ball. 

^^ imputet aliena] aliena imputat Ball. 

^^ Aug. has, after negantis, another explanation introduced by the usual Sine: 
see the text ad loc. 


(paraphrastic) in 1 Cor. vii 14 

exempla iam praecesserant et uirorum exemplum refert quia saepe contigerit 

quos uxores et feminarum quas mariti ut lucri fieret uir* per mulierem .... 

lucri fecerant Christo et paruulorum saepe enim si^ contigerat ut filii 

ad quos faciendos Christianos uolun- illorum ^ parentem qui crediderat 

tas Christiana etiam unius parentis sequerentur. 

Once again we have the important Augustinian quotation com- 
pletely present, as well as the' requisite parallels to the two para- 
phrastic passages. The Reichenau and Balliol MSS also have there- 
fore a claim to be regarded as Pelagius. A marked difference, 
however, between this case and that of the rivals, St Gall 73 and 
Paris 653, lies in the fact that, while the latter two give all that 
Augustine gives, they give a good deal more than that. The 
Reichenau and Balliol MSS would therefore, on general grounds of 
probability, have a better claim than their rivals to represent the 
original, uninterpolated Pelagius. But, meantime, if this were all 
our evidence, we could not be absolutely certain as between the 
different claimants, and it is left to the evidence of Mercator to 
decide the case between these ^ 

A somewhat different method may be adopted on this occasion 
to save space. Having established a probability that the Reichenau 
and Balliol MSS represent the original form of the commentary, 
let us first set out in parallel columns the evidence of Mercator and 
of these MSS. 

Marius Mercator* Reichenau and Balliol MSS 

(References to cod. come from a collation ({n Rom. v 12) 

kindly made by Dr C. H. Turner.) 

propterea sicut 
per unum hominem. peccatum intrauit per unum hominem ^ in hunc mundum, 
in mundum., et per peccatu-m mors. peccatum introiit'', et per peccatum 


^ sic Aug. 7rt2 e.x tur Aug. ml. - sic etiam Ball. ^ ilium recte Ball. 

* The importance of Mercator's evidence in this connexion first became clear to 
me on reading an unprinted paper by Dr Armitage Robinson, written about 1890, 
which he has kindly permitted me to use. 

5 Ed. Baluze, pp. 135 ff., Migne, P.L. xlviii pp. 85 — 87; see also Garnier's 
comparison between Mercator and Pseudo-Jerome, pp. 589 — 593. 

« These five words are given by the Balliol MS at an earlier point. 

7 intrauit cod. Aug. 




Exemplo seu imagine usus est; quia 
sieut, cum non esset peccatum, per 
Adam subintrauit, sic et, cum non 
remansisset iustitia apud aliquem, 
uita per Christum reparata est. 

et in omnes homines mors pertransiit. 

Cum sic qui peccant, similiter et 
moriuntur : neque enim aut in Abra- 
ham aut* Isaac aut* lacob mors per- 
transiit, de quibus dominus ait: 'hi^ 
omnes uiuunt.' hie autem propterea 
dicit omnes mortuos, quoniam mul- 
titudine peccatorum non excipiuntur 
pauci iusti, sicut et ibi inquit: 'non 
est qui faciat bonitatem, non est usque 
ad unum'; et itenim illud inquit: 
'omnis homo mendax.' Aut certe in 
illos omnes pertransiit, qui humano 
I'itu, non caelesti, sunt conuersati. 


1. Exemplo uel forma ; quo 
modo, cum non esset peccatum, per 
Adam aduenit, ita etiam, cum paene 
apud nullum iustitia remansisset, 
per Christum est reuocata. 

Et quo modo per iUius peccatum mors 
intrauit, ita et per huius iustitiam uita 
est reparata^. 

et ita in omnes homines pertransiit^, in 
qxto omnes peccauerunt. 

2. Dum ita peccantes^ similiter 
moriuntur : non enim in Abra- 
ham et Isaac et lacob ^ per- 
transiit, de quibus dicit dominus : 
'omnes enim iUi uiuunt".' hie autem 
ideo dicit omnes mortuos, quia in mul- 
titudine^ peccatorum non excipiimtur* 
pauci iusti, sicut ibi : 'non 
est qui faciat bonum, non est usque 
ad imum' ; et : 

' omnis ^0 homo mendax.' Siue: In 
eos omnes pertransiit, qui humano, 
non" caelesti, ritu^^ uiuebant. 
tisqtie ad legem enim peccatum'^^ in 
hoc^* mundo. 

3. Lex'^ peccati uindex aduenit, ante 
cuius aduentum peccatores liberius 
uel praesentis uitae longitudine frue- 
bantur. erat quidem ante legem 
peccatum, sed non ita putabatur esse 
peccatum, quia iam paene oblitte- 
ratum^^ fuerat in^" natura. peccatum 
autem non imputatur cum lex non est. 

4. Quo modo mors regnauit, si noa 
inputabatur^* peccatum, nisi subau- 
dias : 'in praesenti,' non inputabatur. 

* <y= praeparata est Ball. 2 mors pertransiit cod. Aug. 

^ peccant et cod. Aug. * add. in Gam. & et lacob om. Ball. 

8 huic Garn. 7 de — uiuunt om. Ball. ^ miiltitudinem Ball. 

" non excipiuntur] nuncupantur Ball. i" et omnis om. Ball. 

" non] et non cod. Aug. ^^ more Aug. ^^ peccatum] add erat cod. Aug. 

^■* hoc om. cod. Aug. ^s Lex om. Ball. 

^^ paene obliterata cod. Aug. poen§ oblitterarum Ball. 
1^ in] scientia cod. Aug. is inputatur cod. Aug. 




sed regnauit mors ah Adam usque ad 
Moysem, etiam in eos qiii non pecca- 
^lei'unt^ in similitudinem praeuari- 
cationis Adae. 

Siue : Cum non esset qui inter 
iustum et iniustum discemeret, 
putabat mors se omnium dominari. 
Siue : In eos qui mandatum tam- 
quam Adam praeuaricati sunt : hoc 
est, de filiis Noe, quibus praeceptum 
est ut animam in sanguine non man- 
ducarent ; et de filiis Abraham, quibus 
circumcisio mandata est : sed et in 
eos qui praeter mandatum legem con- 
tempserant naturalem, 

qui est forma futuri. 

Quoniam sicut Adam praeter coitum 
a deo formatus est, sic et Christus ex** 
uirgine, fabricante spiritu sancto, pro- 
cessit. Siue : Sicut quidam dicunt, 
forma a contrario; hoc est, ut* sicut 
ille caput peccati, sic^ iste caput ius- 
titiae est (sit cod.), 
sed non sicut delictum., ita et donum. 
Ne in forma aequalitas putaretur. 
si enim in unius praeuaricatione multi 
mortui sunt, midto magis dono " 
et gratia dei per unum hominem 
Christum in midtos ahundauit. 
Plus ualuit gratia in uiuificando 
quam peccatiim in occidendo, quia 
Adam se^^ solum et^^ suos posteros 
interfecit, Christus uero et eos qui 
tunc erant in corjjore, et eos (hos cod.) 
qui postea futuri erant, liberauit. 
Hi autem, qui contra traducem pec- 
cati sentiunt, aliter^* eos qui defend- 
unt traducem impugnare conantur; 

^ etiam cod. Aug. 

^ praeuaricauerunt Gam. fort. rede. 

^ sanguinem Ball. ^ a Garn. 

* ut ovi. Gam. ^ sic] sic etiam Gam. 

11 donum Gam. ^^ se] non se Garn. 

1* acriter Gam. 

sed regnauit mors ah Adam usque ad 
Mosen, et^ in eos qui^ peccauerunt in 
similitudine^ praeuaricationis Adae. 

5. Siue : Dum non esset qui inter 
iustum et iniustum ante distingueret, 
putabat se omnibus dominari. 

Siue : Non solum in eos qui praecep- 
tum sicut Adam transgressi sunt: 
hoc est, de filiis Noe, quibus iussum 
est ne animam in sanguine^ man- 
ducarent ; et de filiis Habrahae, quibus 
circumcisio mandata est : sed etiam in 
eos qui sine praecepto legem con- 
tempsere naturae. 

qui est forma futuri. 

6. Siue: Ideo forma fuit Christi, 
quia sicut Adam sine coitu 
a deo factus est, ita ille ex 
uirgine, spiritu sancto operaute, pro- 
cessit'. Siue: Ut quidam dicunt 
forma a contrario; hoc est, sicut 
ille peccati caput, ita et iste ius- 
sed non sicut delictxmi, ita et gratia i". 

7. Ne in forma aequalitas putaretur. 
si enim unius delicto midti mortui 
sunt, multo magis gratia dei et donum 
in gratia tmius hominis lesu Christi 
in plures ahundauit. 

8. Plus praeualuit iustitia in uiuifi- 
cando quam peccatum in occidendo, 
quia Adam tantiuu se et suos posteros 
interfecit, Christus autem et qui 
erant tunc in corpore et posteros 

Hi autem, qui contra traducem pec- 
cati sunt, ita illam 

impugnare nituntur : 

qui non cod. Aug. 

* similitudinem cod. Aug. 
^ processit operante Ball. 
I'' donum cod. Aug. 
1^ et] sed et Garn. 




'si peccatum,' inquit^, 'Adae etiam 
non peccantibus nocuit,ergo etChristi 
iustitia non credentibus prodest, quo- 
niam similiter, immo plus, dicit apo- 
stolus per unum liberari quam per 
unum ante perierunt*.' deinde dicunt : 
'si baptismus mundat antiquum illud 
ueternosumque peccatum, qui de duo- 
bus baptizatis nati fueriut, debent hoc 
carere peccato ; non enim potuerunt^ 
ad posteros transmittere quod ipsi 
minime habuerunt".' in hoc ad- 
dunt quoniam, 'si anima non est ex^ 
traduce (sicut nee est), sed sola caro 
habet traducem peccati, 
sola et poenam meretur.' iniustum 
est enim ut bodie nata anima non 
ex massa Adae tarn antiquum pecca- 
tum portet alienum, quia nee ^^ 
rationabile est ut deus, qui pro- 
pria peccata dimittit, unum^^ imputet 

'si Adae,' inquiunt, 'peccatum etiam 
non i^eccantibus nocuit, ergo et Christi 
iustitia etiam non credentibus prodest, 
quia similiter, immo et magis, dicit ^ 
per unum saluari qviam^ per 
imum ante perierant".' deinde aiunt: 
'si baptismum mundat antiquum illud 
delictum, qui de duo- 
bus baptizatis nati fuerint, debent hoc 
carere peccato ; non enim potuerunt 
ad fihos transmittere quod ipsi 
minime habuerunt.' illud quoque ac- 
cidit quia, 'si^ anima non est ex 
traduce, sed sola caro, 

ipsa tantum habet traducem peccati, 
et ipsa sola poenam meretur.' iniustum 
esse diceutes ut hodie nata anima non 
ex massa Adae tarn antiquum pecca- 
tum portet alienum, dicunt etiam" 
nulla ratione concedi ut deus, qui pro- 
pria ^^ peccata i-emittit, imputet 
aliena ^\ 

A study of the contents of these parallel columns is instructive 
in various ways. The evidence of Mercator is more important in 
regard to matter than text. It must be remembered that the 
Commonitorium super nomine Caelestii, from which these quota- 
tions are taken, was composed by its author in Greek, and after- 
wards translated by the author himself into Latin^^ He need not 
have turned up afresh the places in his copy of Pelagius's com- 
mentary, in order to give the ipsismna iierba, and he does not 
appear to have done so. The differences from Pelagius's exact words 
are just such as might readil}^ have emerged in such a process. 

There is only one difficulty about these quotations, and that is 
the short passage in the first note, from 'Et quo modo' down to 
'est reparata.' Either Mercator or his scribes have, intentionally 

1 inquiunt, Gam. recte. 

* perierat Gam. 

"^ habuerint Gam. 

^ quia si] quasi Ball. 

2 decet Ball. 
^ perierunt cod. Aug. 
* ex est Baluze, a misprint. 
'" quia nee] quin et Garii. 

^ qua Ball. 

6 poterunt Garn. 

^1 etiam om. Ball. 

^^ unum] non Gam. ^^ propria] add. homini Aug. ^■^ aliena imputat Ball. 
15 Cf. Teuffel, Gesch. der rom. Lit.« in (Leipz. 1913) § 456 (1) ; Zimmer, pp. 254 f., 
see above, p. 4 n. 3; Schanz, Gesch. d. rom. Litt. iv (2) Miinchen (1920) p. 481. 


or accidentally, omitted these words. They are certainly not of 
material importance, being somewhat of a repetition of the pre- 
ceding sentence : but on the whole it seems more probable that we 
have here to do with a scribal error at some stage or other. 

If the rest of the columns be compared, the result is strongly 
in favour of the Reichenau and Balliol MSS, as representing the 
original form of the commentary. For it will be observed that the 
comments proceed in the identical order in both columns, and that 
in one place where 'pauca' are indeed omitted, Mercator is careful 
to put the words *et post pauca,' showing that there, and there 
alone, he has omitted something that was not germane to his 
purpose. It is here that the rival MSS St Gall 73 and Paris 653 
fail us, because, if Mercator's citations be compared with the com- 
mentary given in either of these, it will be found that 'pauca' are 
omitted by Mercator in more than one place. The same is true of 
the Pseudo-Jerome form, as can be readily seen from Garnier's 
parallel columns^; but it is not necessary for us to take any further 
account of Pseudo-Jerome's claims to be the original Pelagius, 
because these have already been disallowed as the result of our 
examination of the Augustinian quotations. 

In St Gall MS 73 the following parts are found which are absent 
alike from Mercator and from the Reichenau and Balliol MSS. 

At the end of §1 (reparata): — 

futura, non praesens. 

At the end of §2 (uiuebant): — 

Item : Nunc apostolus mortem animae significat, quia Adam 
preuaricans mortuus est^ sicut et propheta dicit: anima quae 
peccat ipsa morietur: transiuit enim et in omnes homines, qui 
per naturalem legem preuaricati sunt. 

In quo omnes peccauerunt. Hoc est: in eo quod omnes pecca- 
uerunt, exemplo Adae peccant. 

At the end of § 3 (in natura) : — 

Item: Dicens 'usque ad legem,' Moysi significat legem: inferens 
autem 'peccatum non inputatur cum lex non est,' naturalem iterum 
ostendit legem, per quam preuaricatus est Cain, et post ipsum qui 
naturalem legem preuaricati sunt. 

At the end of §5 (contempsere naturae): — 

1 Migne, P.L. xlviii pp. 589 — 593. ^ est {or et) above the line. 


Etiam in eos qui non 'peccauerunt in similitudinem praeuari- 
cationis Adae. Hii sunt qui non in similitudine preuaricationis 
Adae peccauerunt, qui per naturalem legem transgressi sunt, et 
non sicut Adam per mandatum. 

At the end of §6 (iste iustitiae): — 

Item : Forma Christi Adam factus est : sicut enim Adam primus 
mandatum dei preuaricans exemplum est legem dei preuaricari 
uolentibus, sic et Christus uoluntatem patris conplens exemplum 
est imitari cupientibus eum. 

At the end of §8 (imputet aliena): — 

Hie manifesto docet quia non generaliter de omni homine dicit 
dicens : ' unius delicto multi mortui sunt,' quia communi et naturali 
morti non solum peccantes sed et iusti moriuntur. 

Thus, leaving out of account the extract last given, there are 
five more instances of 'pauca' in this MS than there are in the 
Reichenau and Balliol MSS, but of these five passages there is not 
a word in Mercator. 

Let us next take the case of Paris MS 653. In it the following 
notes are found, which are present neither in Mercator nor in the 
Reichenau-Balliol group. 

At the end of §1 (reparata): — 

Alitor: Nunc apostolus animae mortem significat, quia Eua 
pi-aeuaricans mortua est, sicut et propheta Ezechiel dicit : ' anima, 
quae peccat, ipsa morietur ' : transiuit enim et in omnes homines, 
qui naturalem legem praeuaricati sunt. (Then follows §2^ pre- 
ceded by its scripture text.) 

At the end of § 2 (uiuebant): — 

exemplo inoboedientiae. 

In quo omnes peccauerunt. In quo, inobedientiae peccato. Sine: 
In Adam omnium peccantium paena est praemonstrata : hoc est, in 
60 quod omnes peccauerunt. 

After 'fruebantur' in § 3: — 

Alitor : Dicens ' usque ad legem,' Mosi significat legem : in- 
ferens autem ' peccatum non imputatur cum lex non est,' naturalem 
iterum ostendit legem, per quam praeuaricatus est Cain, et post 
ipsum qui naturalem legem praeuaricati sunt. 

^ Certain interesting differences of text which separate this form from all other 
authorities, need not be mentioned here, but see pp. 37 ff. 


After Adam in the lemma of §5: — 

'Adam': id est, homo; hominis autem nomen tam uiro quam 
etiam feminae conuenit ; scriptum est enim ' et benedixit illis, et 
uocauit nomen eorum Adam in die qua creati sunt^' 

After Moysen in the lemma of § 5 : — 

Quidam dicunt : ' usque ad finem Moysi ; id est, legis.' 

After the end of the lemma of § 5 : — 

Hi sunt, qui non in similitudine praeuaricationis Adae pec- 
cauerunt, qui per naturalem legem transgressi sunt, et non, sicut 
Adam, per mandatum. 

Alitor : ' Hie est liber generationis Adam, in die qua creauit 
deus hominem, ad similitudinem dei fecit ilium : masculum et 
feminam creauit eos, et benedixit eos. et uocauit nomen eorum 
Adam in die qua creati sunt I' 

Item alitor : De superiore sententia. 

At the end of § 6 (iustitiae) : — 

Alitor: Forma Christi Adam factus est: sicut enim Adam, man- 
datum dei per Euam praeuaricans, exemplum est legem dei prae- 
uaricare uolentibus, sic et Christus, uoluntatem patris conplens, 
exemplum est imitari eum desiderantibus. 

At the end of § 7 (putaretur) : — 

Alitor : Omnis apostoli sensus hie est, ut dicat plus egisse gra- 
tiam per Christum quam per diabulum Aeue subreptum fuisse. 

With reference to multi in the lemma of § 8, 
a marginal note says : non ergo omnes. 

After mortui sunt in the lemma of § 8 : — 

Hie manifesto docet quia non generaliter de omne homine dicit, 
dicens : ' unius delicto multi mortui sunt,' quia communi et naturali 
morte non solum peccantes, sed et iusti moriuntur. 

With reference to plures in the lemma of § 8, 
a marginal note says : quia plures inuenit quos seduceret. 

At the end of the lemma of § 8 : — 
et non, sicut per unum peccantem, ita et donum. 

At the end of § 8 (imputet aliena) : — 

Alitor : Adam solam formam fecit delicti, Christus uero et gratis 
peccata remisit et iustitiae dedit exemplum uiuendi. 

Alitor: Ostendit maiorem uim esse gratiae quo in se homines 
1 Gen. V 2. 2 Gen. v 1—2. 


trahat, quam peccati, id est, diabuli : nam diabulus Aeuara decepit, 
et per exemplum illius ad alios cucurrit, gratia uero et multos in- 
uenit, quos credentes iustificauit, et permultos facile in se alios 

In the case of Paris MS 653, the differences are even more 
glaring than in St Gall MS 73. There are in this MS, leaving out 
of account the extra lemma given above, and the notes at the end 
of § 8, eleven portions, not one of which is present either in Mercator 
or in the Reichenau and Balliol MSS. Hovt^ever interesting and 
even venerable the extra notes in Pseudo-Jerome, St Gall MS 73 
and Paris MS 653 may be, it is now clear that they are no part of 
the original form of Pelagius's cotnmentary, which exists untouched 
in no known documents save the Reichenau and Balliol MSS\ Of 
this discovery further confirmation will be adduced in the sequel. 

The Vatican Fragments 

Mgr Giovanni Mercati discovered in the Vatican Library about 
1905 or 1906 two leaves of a sixth century MS in half-uncial 
writing'^. This MS had become mutilated, and was further broken 
up to be used as guard-leaves at least as early as the eleventh or 
twelfth century. 'The leaves are mutilated, scribbled over and cut 
down, and part of the blame must rest on the nineteenth-century 
binder who separated them unskilfully from some unknown MS or 
printed book, with damage to some letters.' 

'The two leaves are conjugate, but not consecutive^ and contain 
(with lacunae) part of the Pelagian commentary on Rom. vii 9 — 15, 
viii 3 — 8, in a much briefer recension than that published in Migne, (ed. 1846)676 d— 677 D,680 a— d (=702 c— 703 d,706 b— 
707 A of the edition of 1865).' The conclusion to which Mercati 
came as to the length of the gap between the two leaves is con- 

1 The Merton MS 26 is left out of account, as it is a copy of the Balliol MS : 
see pp. 223 ff. 

- See his article in the Journal of Theological Studies vol. viii (1906 — 1907) 
pp. 529 — 535, with a supplementary note by the present writer, pp. 535 f. The words 
in inverted commas are quoted from this article. I have to thank him for rotographs 
of the pages. 

^ In fact, the third and sixth of a quaternion, Mercati on p. 531. 


firmed by the Reichenau and Balliol MSS. The 'recension' is, in 
fact, identical with that in the Reichenau and Balliol MSS, while 
the scriptural lemmata approximate more to those in the latter. 
' It is remarkable that the passages omitted are, as it were, so many 
alternative interpretations introduced by item, all of which can well 
be detached; and when they are detached the residue forms a well 
connected whole, noteworthy alike in style and doctrine. And the 
doctrine is Pelagian at Rom. vii 8.' 

The MS in fact, when complete, must have been closely related 
to a contemporary ancestor of the Balliol MS. For proof of this 
readers are referred to my critical apparatus. 

The arrangement of text and commentary in these venerable 
fragments is a matter of some consequence, and hence one page is 
reproduced here. The reader is referred to Dr Mercati's article for 
the evidence that this arrangement is maintained throughout. 
Letters with a dot under them survive only in part: lost letters are 
in italics. 

(I a) 
(P.L. XXX 702 c fin.) (Rom. vii. 9—13) 

per scientiam naturalem et 
mortuum fuerat per oblim'onem 
ideo dicitur reuixisse per legem 
{v. 10) ego autem mortuus sum qui sci 

(5) ens praeuaricaui. 

et inuentum est mihi manda^wm 
, quod erat in uita hoc esse in morte 

quod custoditum proficiebai ad 
uitam neglectum duxit a.d mortem 
(v. 11) (10) nam peccatum occasione accepta 

per mandatum seduxit me^ et per 
(v. 12) illud occidit me • itaque lex quidem 

sea • et mandatum scm • et iustum 
et bonum • contra inpugnatores 
(15) legis • et contra eos qui iustittam 

a bonitate secernunt • lex • et sea 

1 me cancelled. 
S. P. 4 


et bona dicitur • et gratia iusta 
nisi enim abundauerit iusti^zo 
iiestra • sed et ds • non numqu«7/i 
(20) in ueteri bonus • et in nouo dicitur 

iustus • pater iuste ait dns • h*c con 
tra marcionita55l«**** 
(v. 13) q?iod ergo bo?iz<wi***. 

The meaning of this arrangement is that scripture lemmata were 
always begun a little to the left of the vertical line bounding the 
left edge of the comments, and as a rule the comments throughout 
were bounded by a vertical line a little to the right of that bound- 
ing the lemmata. In this way one could more readily turn to a 
particular passage, the diflference between text and comment being 
giaphically represented. But still more important for our purpose 
is the knowledge that the arrangement was that a short extract 
of scripture should be followed by a short note underneath it, 
by a subnotatio in fact. Pelagius himself at m Col. iii 19, sicut ad 
Ephesios plenius subnotatum est, shows what his arrangement 
was. It is confirmed by the Reichenau MS, the Balliol MS for the 
most part, one family of Pseudo-Jerome MSS, Cassiodorus, etc., 
and though it is quite true that some MSS suggest a 'packing' of 
scripture text, for example in Philippians, that is clearly at variance 
with the general practice, and therefore secondary. Whatever 
others may have done after him, Pelagius did not insert his 
comments between the lines in a manuscript of the Apostle, but 
copied out almost the whole of the Epistles, clause by clause, 
comment by comment, clause and comment alternating throughout. 

The page we have copied from the Roman MS can be used also 
to prove in some detail that the MS when complete, was of the 
same content as the Reichenau and BallioP MSS. Following our 
previous method, and taking each claimant in turn, we have the 
following result. On this occasion we can include Pseudo-Jerome, 
lest any lingering doubts as to its character should remain. 

St Gall MS 73 and Ps.-Hier. add the following:— 

At the end of 1. 9 (ad mortem): — 

Item: Poenae mortem dicit, quam nobis peccatum per decep- 

^ With the slight qualification mentioned in chap, vi p. 212. 


tionem suam adquisiuit, ostendens temporalia et terrena, quae 
putantur esse bona, et legem praeuaricare(-i) suadens [illud pec- 
catum], quae [et] uitam aeternam facientibus [rejpromittit. 

Paris MS 653 adds the following: — 

At the end of 1. 5 (praeuaricaui): — 

Et iam nunc multi sunt Christiani, qui crimina sola putant esse 
peccata: si ea non fecerint uiuere se arbitrantur. sed cum eis 
ostensum fuerit et ilia grauia esse quae faciunt, statim mortuos se 
esse cognoscunt. 

Aliter: (then follows the explanation concluded in 11. 1 — 3 of 
the page of the Roman MS). 

At the end of 1. 9 (ad mortem) : — 

Aliter: Poene mortem dicit, qua nobis peccatum per deceptionem 
suam adquisiuit, ostendens temporalia et terrena, quae putantur 
esse bona, et legem praeuaricari suadens, quae uitam aeternam 
facientibus promittit. 

After 'occidit' in 1. 12, the MS being without 'me': — 

Dum non solum peccato delector,sedetiam mandatum contemno. 

For 11. 14 — 16 (contra — lex et) has: — 

Contra Manicheos, qui uetus testamentum inpugnant. Et lex 

After ' Marcionitas ' in 1. 22: — 

et ceteros (and then 11. 14 — 16, as in Roman text). 

It is hardly necessary to remark that not one of the additions 
in these authorities is to be found either in the Reichenau or in 
the-Balliol MS. 

Interpolation in Certain MSS of Ambrosiaster on 
First and Second Corinthians 

Probably few manuscripts of Ambrosiaster have the genuine 
conclusion to the commentary on First Corinthians and the genuine 
beginning to that on Second Corinthians, but among these are the 
manuscripts of Troyes (432, saec. ix — x), Cologne (xxxiv, saec. x)\ 
Cheltenham (518, saec. xv in. written in the Low Countries), and 
Petrograd (F. v. i No. 17, saec. xi, formerly of St Benignus, Dijon)l 

1 For these MSS see Joiirn. Theol. Stud, iv (1902—1903) p. 90. 

2 Father Brewer, S.J., the future editor, has found some others, not here given. 
Claudius of Turin's copy of Ambrosiaster also was of this kind. 



The bulk of the manuscripts have an interpolation at this point. 
Even the oldest, that of Monte Cassino, 150, written in a semi- 
uncial hand before the year 569, contains the interpolation in place 
of the original text\ In fact the Benedictine editor was the first 
to publish the true text, though he strangely omitted to publish 
the true prologue to Second Corinthians^. What had happened 
was that, at least as early as the middle of the sixth century, an 
anonymous MS of the Ambrosiaster had been accidentally or 
intentionally mutilated at that point. When it became necessary 
to copy that mutilated manuscript, the loss was observed, and was 
made good from another commentary. The resulting composite 
manuscript had a large progeny. What I will proceed to show is 
that the commentary used was an uninterpolated Pelagius*. 

Without attempting to examine all the MSS of Ambrosiaster 
containing the interpolation, I have yet endeavoured to construct 
a critical text of it by the aid of some of the MSS, particularly 
the Monte Cassino MS, and certain MSS at Paris^. The Paris MSS 
in the order of quality are: — 1759 (saec. ix in.), 1761 (saec. ix — x, of 
North-Italian provenance), 13,339 (saec. ix). The second MS is 
the earlier Colbertinus of the Benedictines^ The result of the 
critical process is to give us practically a fragment of another sixth 
century MS of the original form of Pelagius. 

In place of in 1 Cor. xv 44 — in 2 Cor. i 6, Migne P.L. xvii 

1 On this MS, see the literature in Study of Ambrosiaster pp. 14 f., and add 
facss. 53 and 60 in Zangemeister- Wattenbach's Exempla ; 23 c in Steffens' Lateinische 
Paldographie^; Spicilegium Casinense vol. in (2) (1901), complete text; E. A.Lowe, 
Beneventan Script (Oxford, 1914) p. 264. 

2 Published by the present writer from collations of two MSS, lent by Father 
Brewer, in Journ. Theol. Stud, iv (1902—1903) pp. 89—92. 

3 The Benedictine editor had of course observed the connexion with Pseudo- 
Jerome (cf. Migne, xvii 283 d, 284 b, a note on 1 Cor. xv 44) : cf. also Vallarsi (and 
Migne) on Ps.-Hier. ad loc. 

^ The earlier acquired Paris MSS of Ambrosiaster's commentary are accidentally 
omitted from the index to the old catalogue (Paris, 1744), and are thus absent from 
my list in Study of Ambst. pp. 14 ff. I ought, however, to have remembered 
P. Corssen's Epistula ad Galatas (Berol. 1885) p. 36. See the next note. 

5 See ed. Venet. iv (1751) 779 for a list of the MSS used by them. MS 1759 
contains Rom., 1, 2 Cor. ; 1760 (saec. x) contains Rom. ; 1761 contains Rom. (last 
part), 1, 2 Cor., Gal., Eph., Phil., 1, 2 Thess., Col., Tit., 1, 2 Tim., Philem. ; 1762 
(saec. xii) contains Rom.; 1763 (saec. xiii) contains all: ed. Rom. below means 
the Cologne reprint of the Roman edition of Ambrose (t. in [1616] pp. 193 Gff.). 


269 B (284 B of reprint)— 277 A (292 B of reprint) these MSS and 
older editions have the following: — 

Non enim corpus animale semper hie habere potest spiritum 
sanctum, tunc uero, id est in resurrectione,. semper manebit in 
Sanctis. Surget corpus spiritale. Quod possit ire obuiamChristo. Sicut 
scriptuni est: F actus est primus homo Adam in animam. uiuentem,; 
nouissimus Adam in spiritum uiuijicantem. Notandum est quod, 5 
cum duos Adam dicit, eiusdem naturae utrosque demonstrat: quod 
contra Manicheos et Apollinaristas facit, qui negant a dei uerbo 
perfectum hominem esse susceptum. Sed non prius quod spiritale 
est, sed quod animale; deinde quod spiritale. primus homo de terra 
terrenus, secundus homo de caelo caelestis. qualis terrenus, tales et 10 
terreni. Caelestis dicitur, quia non humanae fragilitatis ritu, sed 
diuinae maiestatis nutu et conceptus est et enixus: nam usque 
adeo naturam nostri habuit, ut secundus Adam dicatur et homo. 
Et qualis caelestis, tales et caelestes. Si ideo, ut heretici uolunt, 
nostri generis adsuraptus homo non fuit qui caelestis dicitur, 15 
ergo nee isti naturae nostrae sunt qui caelestes appeliantur: si 
uero de his nemo dubitat, nee de illo est ambigendum. Igitur, 
sicut portauimus imaginem illius ten'eni, portemus et imaginem 
huius caelestis. Peccator imaginem Adae portat; iustus uero ima- 
ginem Christi : ergo, sicut portauimus ueterem hominem ante 20 
baptismum, ita et post baptismum portemus nouum. Hoc auteni 
dico, fratres, quoniam caro et sanguis regnum dei non possidehunt, 
neque corruptio incorruptelam possidebit. Frequenter scriptura car- 
nem pro operibus nominat carnis, utibi: uos autem incarnenon 
estis, sed in spiritu. Aliter: Caro, sicut est, regnum dei non 25 
possidebit nisi inmortalitate uestita. Ecce niysterium dico. Obscuri- 
tatem significat nominando mysterium. Omnes quidem resurgemus, 
non omnes inmutabimur. Omnes autem homines resurgent, sed soli, 
qui regnaturi sunt, in gloriam mutabuntur. Sine: Ita omnes re- 
surgemus, qui in aduentu Christi mortui inueniemur, non omnes 30 
inmutabuntur qui in corpore sint reperti, quia sancti soli beati- 

3. surgit Cas. 7. a om. Cas. 11. quia] qui ed. Rom. 15. dicatur Cas. 

18. portabimus Cas. 26. inmortaliter Cas. mysterium] add. uobis ed. 

Rom. { = vg.). 28. omnes (pr.)] add. autem ed. Rom. autem] enim ed. 

Rom. resurgimus Cas. + sed Cas. 29. resurgimus Cas. 30. inuenimur 

Cas, 31. c^soli sancti ed. Rom. (c. nostro cod. Spinal. Ps.-Hier.). 


tudinis gloriam consequentur. In momento, in ictu oculi. Per ictum 
oculi nimiam breuitatem uult significare momenti, ut quanta sit 
dei potentia, ex resurrectionis celeritate cognoscas. In nouissima 

5^ tuba et mortiii resui^gent incorrupti et nos inmutabimur. Nouissimus 
aduentus intellegiturChristi; mortui autem uel peccatores intelle- 
gendi sunt, qui etiam uiuentes mortui esse dicuntur, ut ad poenam 
aut inmortales aut absque aliqua membrorum diminutione resur- 
gant. uel certe simpliciter omnes mortuos resurgere dicit, et solos 

40 sanctos cum his, qui uiui iusti inuenti fuerint, in gloriam inmutari. 
Oportet enim corruptibile hoc. Necesse est fieri quod promissum est. 
Induere incorruptionem et mortale hoc induere iwrnortalitatem. Est 
quod induit, et est similiter indumentum. Cum autem mortale hoc 
induerit inmortalitatem, tunc fiet sermo, qui scriptus est: Absoi'ta est 

45 mors in uictoria. Ut euacuatis causis mortis per diuinara uictoriam 
ac si absorta non pereat. Ubi est, mors, stimulus tuus? ubi est, mors, 
uictoria tua? Propheta ex persona iustorum loquitur insultantium 
morti. Stimulus autem mortis estpeccatum. Sagitta mortis peccatum, 
per quod animae iugulantur. Virtus uero peccati lex. Dum fortius 

50 et mains fit per scientiam peccatum. Deo autem gratias, qui dedit 
nobis uictoriam per dominuni nostruin lesum Christum. Uictoriam 
illius peccati, in quo lex per carnalem nostram uoluntatem fuerat 
infirmata, quam Christus cruce et exemplo destruxit. Itaque,fr aires 
mei. Reddita resurrectionis ratione, de qua haesitabant, hortatur 

55 COS in dei opere permanere, iam certos de retributione futura. 
Stabiles estate et inmobiles, ahundantes in opere domini semper, sci- 
entes quod labor uesfer non est inanis in domino. Nemo uos de 
gradu spei futurae ultra permoueat. Nam de collectis quae fiunt in 
sanctos, sicut ordinaui ecclesiae{-is) Galatiae, ita et uos facite. De 

60 sumptibus dicit, qui per singulas ecclesias collecti Hierosolymam 
Sanctis pauperibus mittebantur. Per unam sabbati. Una sabbati 

35. tuba] add. canet enim tuba ed. Rom. ( = vg. Clem.) 37. etiam] iam Cas. 

ut ad poenam aut] aut quia ed. Eom. 38. aut (alt.)] aut quia ed. Rom. resur- 

gant] resurgent ad poenam, dicuntur resurgere incorrupti ed. Rom. 40. his] 

iis ed. Rom. 42. iucorruptelam Cas. ed. Rom. ( = vg.). Est] Et ed. Rom. 

43 et] add. quod induitur ed. Rom. 46. pereat cod. un. ed. Rom. pareat codd. cet. 
stimulus] aculeus ed. Rom. {sic 44). 47. insultantium Cas. insultando cett. 

48 CO peccatum est ed. Rom. ( = vg.). 52. nostra ed. Rom. 53. cruce] crucis 

ed. Rom. et om. codd. plerique (add. unus cum Cas.) ed. Rom. 59. ecclesiae 

Cas. al. 61. Una] Unam Cas. 


dominica dies est, sicut in euangelio dicit[ur] dominum una sabbati 
resurrexisse. Umis quisque uestrum apud se recondens quod ei bene 
placuey^it, ut non cum uenero, tunc collectae fiant. Ut paulatim re- 
seruantes non una hora grauari se putent, ut hilares datores dili- 65 
gantur a deo. Cum autem fuero praesens, quoscumque probaueritis 
per epistulam, hos mittam perferre gr^atiam uestram Hierusalem: 
quod si dignum fuerit ut et ego earn, mecum ibunt. Per se clarum est 
quia utrumque in eorum arbitrio derelinquit, ut et quod dederint 
portetur, et per quos direxerint ipsi eant. Ueniain autem ad uos, 70 
ctim Macedoniam pertransiero : nam Macedoniam pertransibo nisi 
uos me duxeritis quocumque iero. nolo enim uos nunc in transitu 
uidere: speroenim me aliquantum temporis mansurum apud uos, si 
dominus permiserit. Quia ita se agunt Macedones, ut non sit 
necesse mihi apud eos diutius remorari: apud uos autem necesse 75 
est ut maneam uel hiemem; multa enim sunt quae corrigantur a 
uobis, sicut medicus ibi moram habet ubi plures aegrotant. Manebo 
autem, Ephesi usque ad pentecosten: ostium autem, mihi apertum, est 
Tnagnum et euidens, sed aduersarii multi. Ideo ibi permanebo, quia 
cum mihi euidens datus sit aditus praedicandi, sunt plurimi qui 80 
resistant. 8i autem uenerit Timotheus, uidete ut sine timore sit 
apud uos, quia opus domini operatur, sicut et ego: nequis ilium 
spernat. Sine tribulationis formidine uel etiam uestri contemptus. 
Deducite aidem ilium in pace, ut ueniat ad me: exspecto enim ilium 
cum fratribus. Nihil admittentes quod ad animi eius proficiat 85 
laesionem. De Apollo autem fratre notum uobis f ado, quia multum 
ilium rogaui uenire ad uos cum fratribus, et utique non fuit uo- 
luntas id nunc ueniret: ueniet autem cum oportunum fuerit. uigi- 
late, state in fide, uiriliter agite et confortamini: omnia uestra in 
caritate fiant. Uigilate mentis oculis ad diaboli astutias praeca- 9° 
uendas: state, quia stantibus difficile somnus obrepit: uiriliter 
62. dicit Cas. ed. Rom. 63. surrexisse Cas. ed. Rom. 65. diliguntur 

ed. Rom. 67. Hier. ] iJraem. in ed. Rom. ( = vg.). 68. et Cas. om. cett. 

ibunt] uenient ed. Rom. 71. Macedoniam alt.] machedonia Cas. nisi — 

duxeritis] Apud uos autem forte manebo, aut etiam hiemabo: ut uos me deducatis 
ed. Rom. {ad \g prope accedit). 72. dux.] dedux. Cas. 75. remorari Ca^. 

remanere cett. autem Cas. om. cett. 11. uobis] nobis ed. Rom. 78. autem] 
om. unus cod.; enim ed. Rom. ( = vg.). 79. quia] qui ed. Rom. 81. resis- 

tunt ed. Rom. 82. apud] inter un. cod. ed. Rom. fort, recte. 87. uoluntas] 

add. eius ed. Rom. 89. et om. Cas. in] cum Cas. et alius, un. cod. ed. Rom. 

91. obrepit Cas. ed. Rom., obripit cett. 


agite; muliebris enim omnis inconstantia et uarietas iudicatur: 
confortamini; ut sit in uestra uirtute prefect us, omnia non inanis 
gloriae causa, sed caritatis gratia facere festinate. Obsecro autem 

95 uos,fratres, nostis domum Stefanae et Fortunati et Achaici, quoniam 
sunt primitiae Achaiae, et in ministerium sanctorum se ordinaue- 
runt, ut et uos suhiecti sitis talibus et omni cooperanti et lahoranti in 
uobis. gaudeo autem in praesentia Stefanae et Fortunati et Achaici, 
quoniam id quod uobis deerat, ipsi adimpleiierunt. Quia praesentes 

loo sunt apud uos et in illis magnum potestis habere profectum. Sine: 
Quia mihi uenerunt pro uobis ministrare officium caritatis. Refe- 
cerunt autem et meum spiritum et iiestrutn. Meum spiritum caritate 
pro uobis, uestrum pro mea laetitia [meum] refecerunt. Cognoscite 
ergo huius modi. Unde et alibi ait: cognoscite eos, qui ita am- 

105 bulant, ut habetis formam nostram. hie 'cognoscite' honorate 
cognoscentes eorum studium uel laborem. Salutant uos ecclesiae 
Asiae. salutant uos in domino Aquila multum et Priscilla cum 
ea quae in domo eorum est ecclesia, apud quos etiam hospitor. Do- 
mesticam congregationem fraternitatis ecclesiam nominauit. Salu- 

ixotant uos fratres omnes; salutate inuicem in osculo sancto. salutatio 
mea manu Pauli. siquis non amat dominum nostrum lesum Chris- 
tum, sit anathema. Sicut illis, qui eum amant, redemptio uenturus 
est Christus, ita qui eum non amant anathemabit; id est, ut illos 
abominetur et perdat. Marana tha. Magis Syrum est quam He- 

1 15 breum, tarn etsi ex confinio utrarumque linguarum aliquid Hebreum 
sonat, et interpretatur 'dominus noster uenit.' Gratia domini nostri 
uobiscum. Propriae manus consueta subscriptio. Caritas mea cum 
omnibus uobis. Ut quo modo uos ego diligo, ita et in Christo in- 
uicem diligatis. In Christo lesu. Non secundum saeculi caritatem. 

120 Atneii. Confirmatio est benedictionis hie sermo, sicut superius ipse 
demonstrat quomodo, inquiens, dicit Amen super tuam be- 

92. muliebris Cas. ed. Rom., mulieria cett. 102. autem] enim ed. Rom. 

(=vg.). Meum — meum] Pro charitate uestrum, pro laetitia meum spiritum ed. 

Rom. Meum spiritum om. un. cod. meume alt. am. Cas. et alii codd. 104. huius 
modi] qui huius modi sunt ed. Rom. (cf. vg.). et om. Cas. ed. Rom. 105. cog- 

noscite] add. id est ed. Rom. 111. nostrumom. ed. Rom. ( = yg.). 113. anathe- 
mabit] anathema sit Cas. ed. Rom. 115. tam etsi ex confinio] tamen ( = Cas.) 
ex sermone ed. Rom. utrumque Cas. 118. ego diligo Cas. et al. cod. ed. 
Ro7u. diligo ego cett. 121. dicet ed. Rom. { — \g). 


Explicit ad Covin thios prima incipit ad eosdem secunda, 
Cuius haec principalis est causa: quoniam in prima pro quo- 
rundam peccatis doctores eorum praecipue corripuerat, et multum 125 
fuerant contristati, nunc eos consolatur, suum eis proponens ex- 
emplum, et docens non debere aegre ferre quod pro aliorum sunt 
salute correpti, cum ipse pro aliena salute periculis cottidie et morti 

Paulus apostohis Ghristi lesu. Quaeritur cur in omnibus epis- 130 
tulis contra usum epistularum primo suum nomen ponat quam 
eorum ad quos litterae destinantur. sed hoc auctoritatis est aposto- 
lici ordinis, qua minoribus scribit, sicut etiam indices saeculi solent 
ad eos quos regunt scripta dirigere. Per uohmtatem dei. Dei, non 
hominum uoluntate: simul ut ostendat non sine patris uoluntate 135 
se missum a Christo. Et Timotheus frater, ecclesiae dei, quae est 
Corinthi. Non dixit: 'Paulus et Timotheus,' quia non ambo apos- 
toli : ad Filippenses uero, ubi non erat tanta auctoritas necessaria, 
'serui' ambo ponuntur. Cuvi Sanctis omnibus. Hie 'sancti' possunt 
accipi sacerdotes, qui in prima ponuntur dominum inuocantes, et 140 
ad Philippenses cum episcopis et diaconis. ideo autem postea no- 
minantur, ne parum intellegentes eos praetermissos esse putarent, 
cum iam sint in ecclesia conprehensi. Qui sunt in uniuersa Achaia. 
Cuius est metropolis Corinthus. Gratia uobis et pax a deo. Gra- 
tias agunt deo, gaudentes se ideo consolari, ut ipsi alios conso- 145 
lentur. Benedictus deus et pater domini nostri lesu Ghristi, pater 
misericordiarum. Quia ex ipso est omnis misericordia. Et deus totius 
consolationis. Id est, perfectae consolationis, quia non est minus 
tribulatione solacium. Qui consolatur nos in pressura nostra. Non 

123 sic Cas. qui add. argumentum. Explicit in Epistolam [primam] ad Corin- 
thios eiusdem Ambrosii I incipit epistola secunda ad corinthios Paris, 1759, 1761 
quorum om. primam 1759 : explicit tractus in epia • I • ad corit Incipit tractat' sci 
Ambrosii inepla • II • adeos, Paris 13339. 124. praem. Secundam Epistolam 

Apostolus scribit Corinthiis ed. Rom. 125. et] unde ed. Rom. 129. subiacet 

ed. Rom. 130. lesu Ghristi ed. Rom. ( = Tg.). 131. ponat] add. Apostolus 

ed. Rom. 132. haec ed. Rom. auctoritatis Cas. auctoritas cett. 133. qua 

Cos. quia cett. etiam] add. et Cas. ed. Rom. 136. dei om. ed. Rom. 

138. necessaria om. ed. Rom. 139. coomnibus Sanctis Cas. ed. Rom. (=vg.). 

141. diacones Cas. 144. et om. Cas. Gratias — consolentur om. ed. Rom. 

145 se] esse Cas. 146. domini nostri lesu Ghristi om. Cas. 147. totius] 

omnis ed. Rom. consolationis {alt.)] add. ut significaret ed. Rom. 149. pres- 

sura] praem. omni ed. Rom. (cf. vg.). 


150 in aliquibus, sed in omnibus. Ut possimus et ipsi consolari eos qui 
sunt in omni angustia, per exhortationem qua exhortamur et ipsi a 
deo. Propter ea liberamur, ut et nos alios consolari et de tristitia 
liberare possimus. Aut: Ita formam nobis dat alios consolandi, ut 
per exhortationem qua ipsi a deo consolaraur, agnoscamus quod 

155 deus timentium se neminem derelinquat, et multo magis in future 
remuneret quos etiam in praesenti non deserit. Quoniam, sicut 
abundant passiones Christi in nobis. Id est, pro nomine Christi. 
Ita et per Christum ahundat etiam consolatio nostra. Ut et Petrus 
est de carcere liberatus, et ipse Paulus uisione domini et uoce con- 

160 firmatus in templo. Sine autem angustiam patimur, pro uestra ex- 
hortatione et salute. Quia uos ad salutem hortamur. Sine: Ut 
nobis exemplum tolerantiae praebeamus. 

I have said above that this is practically a fragment of a sixth- 
century MS of the original form of Pelagius, but it is necessary to 
admit that, if our two authorities, the Reichenau and Balliol manu- 
scripts, contain between them the whole of the original Pelagius, 
then this portion contains a little more than that in the section it 
provides. The following words or clauses are absent from the 
Reichenau or the Balliol MS or both: id est in resurrectione (1. 2), 
Adam (1. 13), ut non cum — ipsi eant (11. 64 — 70) (the most signifi- 
cant of all),wm uos me — permiserit (11. 71 — 74), cum fratribiisi}. 85), 
de Apollo — oportunum fuerit (11. 86 — 88), obsecro autem — in nobis 
(11. 94>— 98), Salutant—Asiae(\\. lOQ— 107 ),apud—hospitor {1 108), 
salutatio — Pauli (11. 110 — 111). The majority of these passages are 
portions of scripture text, which there is some reason to believe 
Pelagius passed over in the course of his commentary\ The others 
must have their claims examined with the rest of the authorities. 
Minor variations between this text and our leading authorities need 
not here be referred to, especially as there are many such between 

150. in aliquibus] in aliqua ed. Rom.; aliquid duo codd. omnibus] omni ed. 

Earn. 151. exhortationem qua exhortamur] consolationem qua cousolamur 

ed. Rom. 152. 00 uos et Cas. ed. Rom. 154. consolationem ed. Rom. agnos- 
camus Cas. ed. Rom. agnoscimus cett. 160. sive] si ed. Rom. angusti»m 
patimur] angustiamur Cas.* (con: m 1). uestri ed. Rom. 161. Quia] praem. 
Id est ed. Rom. exhortamur ed. Rom. 

1 See the evidence ad locos. 


the Reichenau and Balliol MSS themselves, and the Freiburg 
fragments differ at times from both. 

Of greater significance is the absence from this section of pas- 
sages found in all but the Reichenau and Balliol MSS.^ They 
are these: — 

Paris MS 653 

(After 'Sanctis' 1. 3): — 

Aliter: Animale corpus dicit, quod conditum terrae corrumpitur, 
spiritale uero, quod incorruptum resurgit, ut possit aerem penetrare, 
festinans ad caelos. 

Aliter: Primus Adam ad hoc factus est tantum ut uiueret. 

Aliter: Nouissimus Adam, id est Christus, ideo suscepit homi- 
nem, ut uiuificaret. (These three notes are not exactly contiguous.) 

Aliter: Hie primum et secundum hominem iuxta operam ter- 
restrem et caelestem dicit: nam et Christus secundum carnem ex 
nostra massa fuit. 

(After 'possedebit' [sic] 1. 22) in margin : non dixit 'non resurget.' 

(After 'absorta' 1. 44) in margin: in osee propheta. 

(After 'peccatum' 1. 48) in margin: in icto (sic) faciens quasi 

(After 'domino' 1. 57): Hoc est: non inaniter laboratis. 

(Before 1. 58) in margin: incipit de collectis. 

(After 'per' 1. 67) in margin: quasi cum epistulis meis illos 

(After 'signum' [for 'si dignum'] 1. 68) in margin: si non fuerit 
aliqua maior quae me detineat causa. 

(After 'aegrotant' 1. 77): Aliter: Tam diu aput uos ero quam 
diu uestrae placuerit uoluntati. 

(After 'ostium' 1. 78) in margin: ad profectum multorum. 

(After 'resistant' 1. 81): Aliter: Quia et ipse uirtutes faciebat 
et signa et in uirtutibus et in doctrina. 

(After 'fuerit' 1. 88): tunc ei non fuit oportunum. 

(After 'eius modi' [='talibus'] 1. 97): Vel fide primitiae uel 

^ The St Gall MS 73 must be added to these here, because in First Corinthians, 
and there only, it is for the most part free of interpolation. 


(After 'laboranti' 1. 97): qui uel nobis cooperatur uel illis. 
(After 'enim' [='autem'] 1. 102): circa meum obsequium. 
(After 'Christum' 1. 112): non ficto ut iudas. 
(After 'del' 1. 136) in margin: erat ibi et non dei. 
(After 'deo' 1. 144): Patre nostro et dno ihu xpo (=vg.): quod nos 
solemus ille illi salutem. 

Pseudo-Jerome, but not Paris MS 653 

(After 'gloriam consequentur' 1. 32): Aliter: In quibusdamGrecis 
eodicibus habet: omnes enim dormiemus, non omnes muta- 
bimur: in aliis autem: omnes enim non dormiemus, omnes 
autem mutabimur, quod aptat magis ad sensum apostoli, quia 
hie sermo non de omnibus generaliter dicitur nisi de solis Sanctis. 

We have thus evidence of another practically uninterpolated 
copy of the original Pelagius, not later than the sixth century. 

The Cassiodorian Commentary (Pseudo-Primasius) 

The evidence of this commentary and of those that follow is 
not of the same positive character as that furnished by the pre- 
ceding, but it is nevertheless not without significance. The Cassio- 
dorian commentary employs Pelagius so largely (in some Epistles 
being little else than a copy of Pelagius), that it is hard to see 
why not a single interpolated passage of the kind we have been 
making acquaintance with, is found in it, unless it be the case 
that only the uninterpolated form was used by the author. I 
have gone through the whole Cassiodorian commentary, and under- 
lined every borrowing from Pelagius. There is not a trace of a 
single interpolation. It is true that Cassiodorus shows knowledge 
of one or two passages which are absent from the Reichenau MS, 
but these are present in the Balliol MS, which contains a few 
passages absent both from the Reichenau and Roman MSS. 

Cassiodorus thus furnishes additional evidence of the existence of 
a form of the commentar}- such as Augustine, Mercator, the Reiche- 
nau, Balliol, Roman and Ambrosiaster MSS prove to have existed, 
namely one free of interpolation. 

il] how to identify the pelagius commentary 61 

The Extracts from John the Deacon 

One of the numerous discoveries to the credit of Dom Germain 
Morin,O.S.B.,introduces us to the name' John the Deacon' in connex- 
ion with this commentary^ In two manuscripts, Codex latinusmona- 
censis 14,500 (formerly of St Emmeram in Ratisbon) (saec. IX — x), 
and British Museum Harleianus 659 (saec. xiii), he found extracts 
with this name attached to them, which are really by Pelagius. 
As his examination of the British Museum MS was confessedly 
hurried, and the manuscript contains other matters of interest than 
this, it was possible for me to find three other passages in it I 
The long extract from the Munich MS, which I copied in 1913, 
stretches from Rom. vi 3 an ignoratis to Rom. vi 14 paruoli sed 
perfecti. Unfortunately for our purpose, this is a section where no 
interpolations are to be found in Pseudo-Jerome, where in fact 
Pseudo-Jerome differs very slightly from Pelagius. It is not there- 
fore possible to assert that it was the original form of Pelagius which 
passed under the name of lohannes Diaconus, or was used by him. 
Yet there is nothing at all inconsistent with the view that it was the 
pure form he used. 

The Harley MS has the following extracts: — 

(f 13 rb) Humanum quippe indicium multis modis corrumpitur, 
amore, odio, timore: sepe indicium integritate uiolatur et contra 
iusticie regulam interdum misericordia inclinatur. (luditium uero 
dei est secundum iieritatem quia (begins Amb].) 

This is from in Rom. ii 2 — 3, and it is evident that the text has 
incurred some corruption in the course of transmission. 

(f 19 vb) Notandum uero quia recte dicitur redemisse nos, non 
emisse. Ipsius enim per naturam fueramus sed nostris delictis alie- 
nati fuimus^ Si igitur ad peccata non redeamus, fructuosa erit 
nobis redemptio Christi quem Christum deus pater proposuit. 

This is from in Rom. iii 24, and there has been some freedom 
of handling. 

1 Revue Benedictine xxvii (1910) pp. 113 — 17; Etudes, Textes, Decouvertes i 
(Maredsous & Paris, 1913) p. 23. 

2 23 March, 1912. 

3 This is the extract published by Morin, viz. from Notandum to fuimus (or 
sumus, as he gives). 


(f. 24 va) corde • etiam in tribulationibus gloriantur magnitudi- 
nem premii cognoscentes ■ de tribulatione finienda infinitum pre- 
mium acquisituri. 

This is from in Rom. v 3 — 4, and there has again been freedom 
of handling. 

(f 27 vb) (Vel) forma Christi dicitur quia, sicut ille sine coitu 
a deo factus est, ita Christus ex uirgine spiritu sancto operante 
processit. Uel Adam dicitur forma Christi, quia, sicut ille est 
pater omnium secundum fidem et sicut (begins Aug.). 

This is from iJi Rom. v 14, and again there has been some free- 
dom of handling. 

All these four extracts are quite consistent with use of the 
uninterpolated Pelagius. There can be little doubt that they come 
from the Breuiariuni de Sancto Paulo of Johannes Diaconus, of 
which a copy existed in the Benedictine Monastery of Blaubeuren 
in South Germany at the end of the eleventh century, as is proved 
by the catalogue of that library ^ It is not so easy, however, to 
say which among the many persons with the name 'John the 
Deacon' really composed this work, as also the Eocpositum on the 
Heptateuch", and the Breuiatio in Psalmos, which. Dom Morin has 
suggested, with great probability, should be identified with the 
well-known Pseudo- Jerome ^ It is possible that other works should 
also be attributed to the same John^, and we may assign all to the 
sixth century ^ 

It is significant that certain even of the interpolated forms 
witness to the originality of the uninterpolated. Both the St Gall 

^ See G. Becker, Catalogi Bibliothecarum Antiqui (Bonn, 1885) p. 175: my 
attention was called to the fact in 1914 by Dom G. Morin, who had read the entry 
in a proofsheet of Dr P. Lehmann's volume of catalogues of Old German libraries : 
see now Mittelalterliche Bibliothekskataloge Deutschlands und der Schweiz i Bd. 
bearb. v. P. L. (Miinchen, 1918) p. 19 1. 22: 'Breuiarium loannis de S. Paulo.' 

2 Of which there was a copy in the Corbie library (see Becker, Gatal. no. 136 
item 245). The MS is still extant, being Paris, B.N. 12309 (saec. xi) : see also 
Spicilegium Solesmense i (1852) 265 f., 278 ff. 

3 Cf. Revue Benedictine xxv (1908) pp. 88 — 94, Etudes, Textes, Decouvertes, 
t. I pp. 59 f. 

* See Dom Morin, R.B. xxvii (J.910) p. 116 for suggestions. 

5 Cf. the article 'Joannes Diaconus' in Pauly-Wissowa, Real-Encycl. Bd. ix 
(1916) pp. 1806 f.; Manitius, Gesch. lat. Litt. d. Mittelalters i p. 693; Schanz, 
Gesch. d. rom. Litt. § 1241. 


MS and Paris MS 653 are heavily interpolated, but the first is free 
from interpolation in the commentary on First Corinthians^ while 
the second is free from interpolation in the commentaries on First 
Timothy, Second Timothy, Titus and Philemon. 

Later Compilations 

Zmaragdus of St Mihiel, as we have seen, makes many quo- 
tations both from a pure Pelagius, and also from the Cassiodorian 
revision, both under the symbol P= Pelagius. Nowhere have I found 
an interpolated passage. The evidence suggests that Zmaragdus 
possessed an uninterpolated MS which he knew to be Pelagius, in 
addition to the Cassiodorian revision, which also he knew to be a 
form of Pelagius. Doubtless both his copies were anonymous^ 

Sedulius Scottus of Liege used as his leading authority Pelagius 
in its original, uninterpolated form, which was doubtless accessible to 
him as an anonymous work^ It is doubtful whether he really cites 
a commentary on Hebrews under that name. There is no trace of 
interpolation in him, and he is an exact quoter. Of all compilers later 
thanCassiodorus he is the most satisfactory authority for the parts he 
employs. It is quite uncertain whether the H at p. 144 A (=1 Cor. 
vii 36) really refers to the interpolation in Pseudo- Jerome at that 
point; it may refer to some passage of genuine Jerome^ 

This part of the subject need not be pursued farther. The 
evidence is sufficient to show that a number of copies of the origi- 
nal, uninterpolated Pelagius continued to exist, even after inter- 
polation had begun its work. 

* Cf. Zimmer, Pelagius in Irland pp. 246 ff. 

2 The sources of Zmaragdus are indicated at some length in Journ. Theol. Stud. 
vol. IX (1907—1908) pp. 584—597, supplemented by vol. xxiii (1921—1922) pp. 

3 The sources of Sedulius are fully set out in Journ. Theol. Stud. vol. xvin (1916 — 
1917) pp. 184—228. 

■* I have suggested hesitatingly Adv. louin. 1 13 in the last quoted article, p. 228. 



It will be recollected that contemporaries quote only the ex- 
positions of Romans and First Corinthians, and, though no one 
has disputed the unity of the thirteen expositions, a proof that all 
are the production of one author will not be out of place in a work 
like the present. There is no hint anywhere that any one exposition 
circulated by itself, and all the manuscripts, uninterpolated and 
interpolated alike, give expositions of thirteen Epistles, some even 
of fourteen. Our method of proof will be not unlike that which 
found general acceptance in the case of Ambrosiaster^. We shall 
divide our chapter into five sections : (a) cross references from one 
part of the commentary to another; (6) illustrations of method of 
exegesis in general; (c) community of ideas throughout; (d) favourite 
verses of scripture; (e) community of style and language. 

(a) Cross References from one part of the Commentary 


In Rom. ii 8 iam superius dictum est contentiosum hunc esse 
specialiter, qui aliquid contra suam conscientiam nititur defensare. 

This is a reference to: — 

In Rom. i 29 contentio est, ubi non ratione aliquid, sed animi 
pertinacia defenditur, et ubi non tam ueritas quaeritur, sed intentio 

In Gal. V 22 omnium uirtutum prima est caritas, quam in 
quattuor partes diuisibilem adnotauimus. 

This is a reference to: — 

In Gal. V 14 dilectio uel caritas quattuor modis constat: hoc 
est, etc. 

In 1 Thess. v 15: in the latter part of the note a phrase is 
quoted from the earlier part and preceded by the words ut superius 
dictum est. 

^ Study of Ambrosiaster (vol. vii part 4 in this series) (1905). 



In Col. iii 19 numquam rem naturalem hortaretur, nisi conti- 
nentes esse coepissent, sicut ad Ephesios plenius subnotatum est. 

This is a reference to: — 

In Eph. V 22 quia non eos sicut incontinentes Corinthios lacte 
alebat, sed perfecto continentiae cibo nutriebat. timet enim ne, 
cessante in plerisque carnis officio, aut in mulieribus subiectio aut 
in uiris cessaret caritatis affectus, et non tam continentiam quam 
diuortium docuisse uideretur. ceterum quale est, ut nouae uitae 
praedicator, nulla exsistente causa, hoc doceret, quod naturaliter 
possidebant ! 

(b) Illustrations of Method of Exegesis in General 

One of the most common features, remarked upon by St 
Augustine ^ is the provision of alternative explanations. These 
alternative explanations are introduced almost invariably by the 
word Sine. The following list of references ought to be approxi- 
mately complete: — 

Rom. i 8, 11, 17; ii 2o, 26, 27; iii 11, 15, 20, 21; iiii 11 bis, 13, 
15; V 6, 14 quater; vi 6, 9, 14, 19; vii 9, 15; viii 2, 3, 19 bis; viiii 
16, 17 bis; x 8, 19; xi 15 bis, 29, 34; xii 8, 18; xiii 5, 6, 13; xiiii 
16; XV 5, 7, 15, 24, 26, 29, 31 bis; xvi I. 

1 Cor. i 23; ii 6, 7, 15; iii 12, 13, 16, 17; iiii 9; v 2, 4, 5 bis; 
vi2; vii 17,18,28; viiii 21; x 12; xi 4, 10, 16, 18, 19, 26; xiiii 33 6t5, 
36; XV 3, 7, 51; xvi 17. 

2 Cor. i 6, 7, 11, 16; ii 3 bis, 5, 14; iii 2, 3, 18; iiii 6, 17; v 8; 
vi 10, 13, 15; vii 2, 3, 4 bis, 5, 10; viii 22 bis, (24); viiii 5 ter, 8 bis, 
9 bis, 10, 13; x 7; xi 5; xii 4 bis; xiii 4. 

Gal. ii 16, 19; iii 19; vi 6 bis. 

Eph. i 4; iii 1, 6, 7, 9; iiii 5, 10 bis, 12. 14; vi 4 bis, 24 bis. 
Phil, i 5, 6, 7, 8 bis, 25, 27; ii 5, 6, 17 bis, 18; iii 13, 21; 
iiii 5, 15. 

1 Thess. i 5 bis; ii 13 bis, 16; iiii 6; v 17, 18, 19, 22 bis. 

2 Thess. ii 16. 

Col. ii 18; iii 5, 17, 25 bis; iiii 10, 18. 

1 Tim. ii 2, 7; iii 3 bis, 9; iiii 10; v. 8, 24 quater; vi 4, 6, 9, 12, 20. 

1 De pecc. mer. et rem. iii 4 §9 (C.S.E.L. lx p. 135 1. 7). 


2 Tim. i 1, 12, 14; ii 1—2 his, 14, 18; iii 6. 10, 13, 15 his\ iiii 8. 

Tit. i 6, 9; ii 5, 7, 10; iii 3, 5 his, 15. 

Philem. 6, 14. 

Occasionally an alternative explanation is introduced by Aliter. 
The following instances occur: Rom. iii 4; iiii 2, 4; vi 22; xiii 1 ; 
XV 17; 1 Cor. iii 18; vii 28; viiii 22; xv oO; 2 Cor. v 16; (viii 24); 
viiii 12, 13; xii 5; Phil, ii 2; Col. ii 23. The employment oi Item to 
introduce an additional note, is the exclusive peculiarity of Pseudo- 
Jerome MSS, and it comes before interpolated notes only. 

A favourite type of note is that which attempts to make the 
meaning of St Paul's words clearer, by pointing out what the 
Apostle is not referring to; this might be called the negative 
method of exegesis. An illustration or two will show what is 
meant: — 

Rom. v 9 In sanguine ipsius Xon animalium sanguine, 

sicut in lege. 

2 Tim. i 1 per uoluntatem dei, Non meis meritis. 

Tit. i 1 Paulus seruns dei. Non peccati. 

Other instances of this type of note are: — 

Rom.i4, 8; iiiil7; vlO; vil7,23; vii 7, 18; viii 32; xil,30; 
xii 6, 11; xiiii 15; xv 1.5. 

1 Cor. iii 1, 9, 10; iiii i. 

2 Cor. i 4; v 21; vi 4 his\ vii 1, 9, 11 quater; viii 5; x 12 his; 
xi 6; xii 10 quater; xiii 12. 

Gal. i 1, 4, 14, 15; ii 2 his, 20 his; iii 8, 18; vi 14. 
Eph. i 1 his, 3 hrs, 4, 5 his, 11; ii 8; iii 3, 4 his, 7, 12; iiii 11 his, 
29; v 19, 21; vi 3, 7, 13, 14. 

Phil, i 4, 1(5, 25; ii 12; iii 1, 5, 15; iiii 3, 6. 

1 Thess. ii 2; iii 3, 13. 

2 Thess. iii 12, 13. 
Col. i 6; iii 15 his. 

1 Tim. i 1, 2; iii 1, 12; v 17; vi 11, 17. 

2 Tim. i 1, 12; ii 9; iii 2. 
Tit. i 1, 4; ii 5; iii 1, 2 his, 7. 
Philem. 23. 

No extra-canonical writers are referred to by name, unless they 
happen to be also heretics. Others, whose views the author con- 
siders and usually ends in rejecting, are indicated by the vague 


word quidam (plural). An attempt will be made in a later chapter 
to discover some of the writers thus intended, but at this stage it 
will be enough to give a list of the references to quidam. (In one 
or two cases other writers are referred to as multi or diuersi.) 

Rom. ii 21; iii 28; iiii 8 (quidam, then alii); v 14; viii 3, 19 
(diuersi, quidam); viiii 16, 17 (diuersi), 20 bis, (21 (eos, qui)), 
(26 (eos, qui)): x 5; xi 20, 26; xiiii 2, 15. 

1 Cor. ii 9; viii 1; viiii 22; xi 21, 27; xiiii 19; xv 28 (multi, 
quidam, alii, multi). 35 (multi). 

2 Cor. ii 15; iii 6^; v 13; vii 11; viii 22; viiii 2 (quidam, alii); 
xii 7. 

Gal. iii 19; v 12. 

Eph. i 10 (multi, quidam, alii, alii); ii 2 (multi); iii 18; v 31 ; vi 5. 

Phil, ii 5 (multi). 

Col. ii 14. 

2 Tim. ii 20. 

Another practice followed throughout is the refutation of 
particular heresies from the passages under consideration. Some- 
times heretics in general are referred to, but far oftener the 
individual heretic is attacked by name. The references to passages 
will be found in the index of pi'oper names; here it may suffice to 
give the names, and the number of occurrences of each: Marcion, 
or the Marcionites, appears twice; the Manicheans eleven times; 
the Arians fourteen times; the Photinians five times; the Nova- 
tians four times; the Jovinianists^ four times; Apollinaris twice; 
the Macedonians once. Heretics in general are referred to twenty 
times in all. The allusions are spread equally over the whole work, 
and the method of allusion is very uniform throughout, as reference 
to a later section of this chapter will show^ 

It is this commentator's practice to pay regard to the different 
sections of the Epistles, and to call attention, for instance, to the 
point at which the discussion of a particular topic ceases. For 
instance, after the note on Rom. i 7 occur the words: hue usque 
praefatio'*; at Rom. x 17 occur the words Hinc responsio apostoli; 

- V 9 in Ps.-Hier. addition has quidam. 

- On the bearing these references have on the date of the commentary see chap, i 
p. 4 n. 6. 

^ Under calumnior, contra (p. 86). 

■* Probably genuine, though lacking in the Reichenau MS. 



at 1 Cor, vii 1 occur the words Incipit de coniugiis; at vii 38 con- 
cludit uirginum causam; at Eph. iii 21 hue usque de mysterio 
incarnationis Christi...hwc incipit moralia omni ecclesiae tradere 
instituta; at 1 Tim. i 18 Hue usque de statu suo...hinc dat. There 
are many more instances of the kind, of which the commentary 
on First Corinthians furnishes a goodly number\ 

Perhaps in the interests of* brevity, the author habitually 
brings his short notes into grammatical connexion with the verses 
or clauses commented on. In fact he carries the process so far as 
sometimes to bring the ' argumentum ' into subordination to the 
title, e.g. incipit ad Galatas arguTnentum, quos pseudo-apostoli, etc. ; 
incipit ad Colossenses, quorum auditam fidem, etc.; incipit ad Tituni 
argumentum, discipulum suum episcopum, quem commonet, etc.; 
incipit ad Philemonem, cui apostolus a Roma scribit. Examples 
from the commentaries proper are: Rom. iiii 5 secundum proposi- 
tum [gratiae] dei, quo proposuit gratis... dimittere; 1 Cor. vi 19 
quem habetis a deo, cui grauissimani iniuriam, facitis fornicando; 
2 Cor. i 23 in animam meam, cuius secreta solus agnoscit; Gal. ii 10 
ut pauperum memores essemus, qui omnia sua distrahentes...uel: 
quorum honafuerant a ludaeis inuasa; Eph. iii 7 secundum opera- 
tionem uirtutis eius, cuius uirtus me confirmauit, sine: cuius 
uirtutes, etc.; Phil, ii 30 tradens animam suam: in manus inimi- 
corum; 1 Thess. iii 11 ipse autem deus.,,dirigat uiam nostram ad 
uos, remotis diabolicis scandalis, quibus noster impeditur aduentus; 
2 Thess. i 7 cum angelis uirtutis eius, qui uenient uindicare: Col. 
i 6 et crescit, in numero uel uirtute; 1 Tim. ii 9 similiter et mu- 
lieres, in omnibus quae dixi de uiris; 2 Tim. iii 14 et credita sunt 
tibi, a deo per nos; Tit. i 14 auersantium se a ueritate, noui scilicet 
testamenti; Philem. 5 fidem tuam et caritixtem, quae operibusinno- 
tescit. These examples have been chosen at random, one from each 
commentary, out of hundreds which might have been adduced. 

The separation between the various parts of this chapter is to 
some extent artificial, and thus some of the later arguments might 
very well have come in at this point. 

1 This specialty of the commentary has been noticed by De Bruj-ne, Revue 
B6n6dictlne xxiv (1907) p. 261 : see also below, under Incipio and causa. 


(c) Community of Ideas throughout 

On this subject a whole book might easily be written, and 
doubtless will be written, but it must come from a theolosfian, 
I will not seek in any way to forestall his task by a treatment 
which must necessarily be defective. I will merely select a few 
themes which have struck me in the course of reading. Readers 
who seek a larger treatment will find it in the works of Klasen^ 
and Loofs-, from which they will eliminate such conclusions as 
depend on au interpolated and faulty text. 

No subject occurs with more persistence than the influence 
of example on conduct. The author is never weary of referring 
especially to the force of the Apostle's good example in the lives 
of his converts. The word used is exemplum, but sometimes /or/na^ 
The following list of instances must be fairly complete: arg. omn. 
epist. his; Rom. i 1, 10; iii 21 ; iiii 4, 24; v 1, 12, 16, 19; vi 11, 14, 
18; viii 8, 4, 32; xii 3, 13, 15; xiii 13; xiiii 6, 13, 15, 22; xv 2, 
16 his, 27; xvi 5, 15, 25; 1 Cor. i 26, 28; ii 1; iii 15 his, 17; 
iiii 6, 17 his; v 6; vi 2; vii 3, 14, 25; viii 1, 11; viiii 1, 6, 7, 13; 
X 1, 22, 24, 32, 33; xiii 11; xiiii 6; xv 13, 32, 36, 57; 2 Cor. i 6; 
iiii 5; v 16 his, 18; vi 3, -13 his, 14; vii 2; viii 1, 10, 15; C^al. 
iiii 16; Eph. ii 7; v 27; Phil, i 11, 14, 24; (ii 7), 8, 12, 16, 19, 
21; iiii 7, 9; 1 Thess. arg. i 5, 7 his, 8, 10; ii 1, 14; iii 9; iiii 7; 
V 12; 2 Thess. iii 1; Col. i 23; ii 6; iii 13, 21; 1 Tim. i 16; ii 6, 
15; iii 2 {q\s.o forma), 3; iiii 12 his (a\so forma), 15, 16; v 1, 9, 11, 
13, 22, 23; 2 Tim. i 4, 10, 16; ii 1—2, 15; Tit. ii 3; iii 3; Philem. 
arg. One or two of the examples may be quoted in full to show 
their character: 1 Cor. viii 11 potest et iliis did qui destruunt 
exemplo quod aedificant iierho; 1 Cor. xiiii 6 suum illis proponit 
exemplum; 2 Cor. v 16 nullius ueterum imitamur exemplum... 
quia eis exempla ueterum proponehant; 2 Cor. vi 14 ostendit nemi- 
n em posse et iustwm esse pariter etiniu stum, ualde contraria exem'plsi 

The author frequently states that we are saved gratuitously 
(gratis), e.g. Rom. i 7; iii 21, 24; iiii 4, 5, (6); viii 29; xi 6; 2 Cor. i 12; 

' Theologische QuartaUchrift Bd. lxvii (1885) pp. 244 — 317, especially the 
latter part. 

2 Article 'Pelagius' in Herzog-Hauck's Real-encijklopddie. 

3 Further examples oi forma later in this chapter, p, 100. 


Gal. V 4 ; Eph. i 9 ; 1 Tim. i 2 ; 2 Tim. i 9 : similarly that we are 
saved by the grace of God, not by our own merits; arg. omn. epist. 
dei se gratia, non suis mentis, esse saluatos; cf. Rom. v 1 nemo suo 
merito, sed omnes aequaliter dei gratia sunt saluati. 

He reiterates St Paul's teaching that we are justified by faith 
alone {sola fides) : Rom. i 17; iii 28 bis; iiii 3, 5, ]1; v 1; viii 29; 
xi 25; 1 Cor. vi 10; viiii 21; 2 Cor. v 19; Gal. i 3, 12; ii 2, 14, 17, 
20 ; iii 5, 6, 10, 11, 12, 14, 22, 26 ; v 11, 24 ; vi 16 ; Eph. ii 8, 16 ; 
iii 11; V 5: cf Phil, iii 3, 9; liii 1. 

Foreknowing and predestinating are identified (Rom. viii 29), 
and God called those, who he had foreknown (praescierat) would 
believe. This latter view the author may have got from Ambrosi- 
aster: in any case the teaching occurs fi-equently in Rom., and also 
later. Examples are Rom. viii 29, 30; viiii 10, 15, 27; xi 2; Gal. 
i 15; Tit. i 3. 

The author is intensely interested in moral progress (pi'oficio, 
profectus^), and constantly refers to it. He speaks often of the 
apostle encouraging (prouocare) his converts to show it. A few 
instances are Rom. i 8; xv 14; 1 Cor. i 4; xi 2; 2 Cor. vi 11; 
vii 13, 14; Gal. iiii 21; Phil, iiii 18; 1 Thess. arg.; ii 20; iiii 1; 
2 Thess. iii 1 ; 1 Tim. iii 2 ; 2 Tim. i 16. Gal. iiii 21 may be quoted 
in illustration : detrimentum discipulorum confusio est magistri, 
sicut profectus eorum est glona praeceptoris. This last is a very 
favourite theme, the joy that the progress of the pupil gives to the 

A few minor illustrations of the unity of the commentary in 
this matter may be added; others may be divined from the 
index of scripture passages or of proper names, such as the refer- 
ences to Ananias and Sapphira, Simon Magus, the call of Barnabas 
and Saul'-. 

The connexion between 1 Thess. ii 14 and Hebr. x 34 is twice 
stated: arg. omn. epist., De Hehraeis iiero quid dicendum est, quorum 
Thessalonicenses, qui plurimum laudati sunt, imitatores facti esse 
dicuntur, sicut ipse ait: et uos fratres imitatores facti estis ecclesi- 
arum dei, quae sunt in ludaea ; eadem enim passi estis uos a 
contribulibus uestris, quae et illi a ludaeis ? aput ipsos quoque 

1 See examples of these words under section (e). 

2 See below, (<f) p. 77. 


Hebraeos eadem commemorat dicens: nam et uinctis conpassi estis 
et rapinam bonorum uestrorum cuni gaudio suscepistis, cogno- 
scentes nos habere meliorem et manentera substantiam (Hebr. 
X 34); which ought to be compared with the note on 1 Thess. ii 14 
uos autem, fratres, imitatores facti estis ecclesiarum dei, quae sunt 
in ludaea in Christo lesu. Quibas dicitur: nam et uinctis conpassi 
estis et rapinam bonorum uestrorum cum gaudio suscepistis (Hebr. 
X 34). Quoniam eadem passi estis et uos a contribulibus uestris, 
sicut et ipsi a ludaeis. I know no other independent commentator 
who brings these two passages together. 

The relationship between the Acts of the Apostles and the 
Epistles is stated in similar language in two passages: 1 Cor. xv 32 
niulta dicuntur in Epistidis, quae in Actibus non tenentur, et rtiultu 
in Actibus, quae in Epistulis non scribuntur; 2 Cor. xi 24 haec in 
Actibus non omnia repperiuntur, quia nee in Epistulis omnia quae 
ibi scripta. sunt, continentur- 

The change in the Apostle Matthew's career is stated in almost 
the same words in two passages: Col. iiii 14 (Lucas) ex medico erat, 
sicut Matheus iam apostolus adhuc dicitur publicanus; 1 Tim. i 15 
hie 'sum' pro 'fui' ponitur, sicut Mattheus dicitur publicanus, cum 
iam Christi esset apostolus. 

'Teaching,' etc. are often described as being that of the New 
Testament, or of the Old and New Testament together. Thus, 
compare the following passages: Rom. vii 12 deus non numquam. 
in Vetere 'bonus' et in Nouo dicitur 'iustus'; Rom. x 8 uerbum 
f[^(\Q\...Noui scilicet Testamenti; Rom. xv 14 repleti omni scientia. 
Hoc est, et Noui et Veteris Testamenti; 1 Cor. i 5 in omni uerbo 
et in omni scientia. Id est, tarn Noui quam Veteris Testamenti: 
1 Cor. vii 5; Gal. iiii 24 duo testamenta. Vetus et Nouuni singulos 
populos generantia; Gal. v 23 qui Nouuvi implet, non est sub Vetere 
Testainento; Col. iiii 6 sermo uester in gratia. In Nouo scilicet 
Testamento; Col. iiii 11 hi soli sunt adiutores in regno dei, qui 
mihi fuerunt in solacio. In Nouo Testamento, per quod intratur ad 
regnum; 1 Tim. vi 3 doctrinae. Noui Testamenti; Veteris enim 
erat seuera doctrina; Tit. i 14 a ueritate. Noui scilicet Testa- 
menti; Tit. iii 6 quem efFudit in nos ah\indQ...Abundantius quam 
in Vetere Testamento. 

Twice he reminds his readers that there can be only one 


episcopus in each ciuitas: Phil, i 2 hie episcoipos presbyteros intelle- 
gimus: non enim in una ciuitate plures episcopi esse potuissent, 
sed hoc etiam in Apostolorum Actibus inuenitur; 1 Tim. iii 8 
quaeritur cur de presbyteris nullam fecerit meniionem, sed etiam 
ipsos in episcoporum nomine conprehendit, quia secundus, immo 
paene unus est gradus, sicut ad Philippenses episcopis et diaconis 
scribit, cum una ciuitas plures episcopos habei'e non possit, et in 
Actibus presbyteros ecclesiae iturus Hierosolymis congregauit, 
qiiihus inter cetera ait: (follows Act. xx 28)... episcopos — 

There are several references to the Laity in the commentary, 
perhaps not unconnected with the fact that the writer himself was 
a layman: 2 Cor. viiii 2 quidam dicunt eum superius de laicis 
Sanctis diocisse, hie uero de sacerdotihus Sanctis; 2 Cor. xiii 11 laicos 
iuhet se inuicem exhoHari; 1 Thess. iiii 18 consolamini inuicem 
in uerbis istis.] Notandum quod laicis hoc praecepit, ut alter- 
utrum se doctrinis suis instituant; 1 -Thess. v 11 consolamini 

inuicem et aedificate alterutrum ] Notandum quia hoc 

laicis praecepit, quos etiam monet praepositis exhibere offida can- 
tatis; Col. iii 16 Et hie ostenditur uerbum Christi non sufficienter, 
sed. abundanter etiam laicos habere debere et docere se inuicem etc.; 
1 Tim. iii 1 fidelis sermo: siquis episcopatum desiderat.] 
Prius laicos instituit, de quibus optinii quiqiie in sacerdotium alle- 
gnntur, et sic dicit quales debeant ordinari; 1 Tim. iii 2 unius 
u xoris uirum.] Si illi nee hoc licet, quod uel laico conceditur, midto 
magis ilia quae etiam in laico jyrohihentur; 1 Tim. v 19 iniustum 
est etiam aduersus laicum accusationem recipere, cum hoc nee sae- 
cidares indices faciant: quanto magis aduersus domini sacerdotem ! 

The Apostles had the gift of knowing what was taking place 
at a distance: 1 Cor. i 11 hanc gratiam habebat (apostolus), ut 
nosset absens quid in singulis ecclesiis ageretur, sicut dicit ad 
Colossensis (ii 5) etc.; Col. ii 5 habebant hanc gratiam apostoli, 
ut alibi positi, quid alibi ageretur agnoscerent, sicut Helisaei 
spiritus cum Giezi fuit in uia (cf. 4 Regn. v 26). 

Note the identical definitions of ecclesia, with reference to 
a 'house church': Rom. xvi 5 ostendit congregationem Jidelium 
ecclesiam nominari; 1 Cor. xvi 19 domesticam congregationem 
fraternitatis ecclesiam nominauit. 

This section may be concluded by reference to the illustrations 


from medicine, employed by the commentator. Our writer is not, 
of course, unique in this respect^ but they are sufficiently charac- 
teristic to deserve collection : Rom. v 4 cum etiam midti propter 
spem paruae salutis et curam corporis maximos sustinuerint crucia- 
tus, nee tamen perfectam potuerint consequi sanitatem, quae etiam 
si prouenerit, paulo post morte intercedente soluetur; Rom. v 6 
peccatorum et scelerum languoribus premeremur; Rom. viiii 17 
tale est hoc quod in Pharaone gestum est, quale si medicus de 
cruciatu iatn damnati rei multis inueniat sanitatem, causas inqui- 
rendo morhorum; Rom. xii 15 non dolemus de uno memhro praeciso; 
1 Cor. i 21 alia illis medicina succurritur ; 1 Cor. vii 3 concedatur 
remedium, languore incontinentiae reclamanti non denegetur 
remediuni nuptiarum, quo mode si peritus medicus inquieto aegro 
et neganti se posse a pomis omnibus abstinere, saltern minus perni- 
ciosa concedat; 1 Cor. xi 31 huic causae humana potest succurrere 
medicina; 1 Cor. xvi 6 multa sunt quae corrigantur in nobis; sicut 
medicus ibi moram habet, ubi plures aegrotant; 2 Cor. i 9 omne 
humanum auxilium defecisse et ab illo solo sperandmn esse reme- 
dium, cui etiam mortuns suscitare possibile est; 2 Cor. ii 2 si con- 
tristatur, intellegit se peccasse; sic aeger qui dolorem sentit, potest 
percipere sanitatem et ad inedici laetitiam pertinere; 2 Cor. iiii 4 
dari permittet, quia credere noluerunt; quia et medicus, si inoboe- 
dientem aegrum deserat, ipse ei aegritudinem dicitur prolongasse, 
cum tota illius cidpa sit, qui audire contempsit: tamen, si rui^sum 
roget et obtemperet, potest recipere sanitatem ; 2 Cor. iiii 7 fragili 
cor pore..., in quo etiam cum alios sanemus, ipsi aliquotiens infirma- 
mur; 2 Cor. vii 9 quasi si dicat medicus: 'etiam si doluerit mihi 
tam ardenti me usum esse cauterio, sed nunc gaudeo, non quia 

' Long ago my brother, W. Clark Souter, M.D., remarked to me that Livy was 
fond of medical metaphors. Philo 'reveals a quite definite interest in medicine' 
(H. A. A. Kennedy, Philo's Contribution to Relicjion (London, 1919) p. 13). See also 
A. V. Harnack, Medicinisches aus der Ultesten Kirchenrjeschichte (Leipzig, 1892). 
Of Seneca the Younger, Mr J. D. Duff says : ' Metaphors and similes from medicine 
abound in his writings' (note on ad Helu. 19 §2). On Tertullian and medicine, 
see T. E. Glover, Conflict of Religions in the Early Roman Empire (London, 1909) 
p. 309, n. 3 ; Hoppe, Syntax u. Stil des Tertullian (Leipzig, 1903) pp. 217 ff. Recently 
Prof. A. S. Pease has published an admirable paper, 'Medical Allusions in the 
Works of St Jerome' (Harvard Studies in Classical Philology vol. xxv [1914] pp. 
73—86). On Fulgentius, see 0. Friebel F. d. Mythograph u. Bischof{Fa.devh. 1911) 
pp. 127 ff. 


doluistis, sed quia dolor ille nobis profuit ad salutem'; 2 Cor. vii 14 
quasi peritus medicus agit, qui uulnus iam prope sanatum hlandis 
unctionihus fouet, ut facilius cauterii ustura sanetur; Phil, i 10 
nemo aeger ante sanatur quam quo modo sanari possit agnoscat; 
1 Thess. ii 5 sicut qui hominibus placent, tamquam aegris desiderata 
omnia concedentes et mortem eormn neglegentes; 1 Thess. v 14 ^?-o 
diuersitate morboruni diuersitas adhibenda est medicinae; 2 Thess. 
iii 15 aeger curandus est, non necandus; 1 Tim. v 23 siinul ostendit 
posse quasdam infirmitates creaturarum medicina sanari; 1 Tim. 
VI 4 nee enim recusando fidem penitus moritur, nee ad ueritatis 
scientiam conualescit, sed diuersis accessionibiis^ languet; 1 Tim. 
vi 17 principalem eorum tetigit niorbiim; 2 Tim. ii 17 'cancer' esse 
dicitur uidnus quod in mammillis nascitur feminarum, quibus nisi 
cito subuentum fuerit, cum virus ad cor serpendo peruenerit, nullum 
ultra remedimn est. ita et haereticorum sunt uitanda conloquia, ne 
per aiires inremediabiliter uulnerent mentes; Tit. i 9 'doctrina sana' 
quae sanat audientes; Tit. ii 8 nullius adulationis access ione^ 

(d) Favourite Verses of Scripture 

In this section are included all cases where a verse or portion 

of scripture is quoted or alluded to three times or oftener. 
One verse is quoted or alluded to ten times : — 
Act. V 41 apostoli ibant gaudentes quia pro nomine domini digni 

habiti sunt contumeliam pati (in Phil, i 28): cf. in Rom. v 4; viii 

17; in 2 Cor. i 5; viii 2; in Phil, iiii 6; in 1 Thess. i 6; in 2 Thess. 

i G; iii 5; in Col. i 11. 

One verse is quoted or alluded to eight times: — 

Phil, ii 7 exinaniuit se ipsumformam serui accipiens etc. (in Rom. 

i 1): cf. in Rom. viii 33 — 4; in 1 Cor. iii 23; xi 3; in 2 Cor. iiii 

5; in Eph. i 21; iiii 9; in Phil, ii 5. 

Two verses are quoted or alluded to seven times: — 

1 Cor. viii 10 in idolio recumbentem. . .aedificabitur ad mandu- 

candum idolothyta (in Rom. xv 2): cf. in 1 Cor. viii 1 ; x 7, 13, 25 ; 

in 2 Cor. vi 14; in 1 Tim. iii 3. 

1 Tim. i 9 lex... est data. . .peccatoribus, etc.: cf in Rom. iiii 15; 

viii 2; in 1 Cor. viiii ^0; in Gal. i 4; ii 19; iii 27; v 18. 
' See Thes. Ling. Lat. s. v. accessio III. 


Five verses are quoted or alluded to six times: — 

Math. V 44 diligite inimicos uestros, etc. : cf. in Rom. viii 9 ; in 

1 Cor. vi 8; in Gal. v 14; in Col. iii 13, 15; in 1 Tim. ii 1. 

Math, xiii 22 sollicitudo saeculi istius et uoluntas diuitiaruin, 
etc.: cf. in 1 Cor. vii 26, 28, 33, 34; in Eph. iiii 18; in 1 Tim. vi 9. 

2 Cor. xii 2 in corpore, etc. : cf. in Rom. v 15; in 1 Cor. xv 51; 
in Gal. ii 11; in 1 Thess. iiii 15; v 10; in 2 Tim. iiii 5. 

1 loh. iii 2 nondum appar'uit quid erimus : scimus quoniarn cum 
apparuerit, similes ei (illi) erimus: in Rom. v 11; viii 17, 18; in 

2 Cor. iiii 11; in 2 Thess. ii 14; in Col. iii 4^. 

1 loh. iiii 18 perfecta caritas foras mittit timorem: cf in Rom. 
V 5; viii 31; xiii 7; in 2 Cor. xii 15; in Gal. v 6; in 1 Thess. i 3. 

Three verses are quoted or alluded to five times : — 

Esai. vii 9 nisi credideritis, nee intellegetis: in Rom. xi 8; cf in 

1 Cor. i 24; in 2 Cor. iii 14; in 1 Thess. ii 16; in 1 Tim. i 18. 
Phil, ii 3 nihil per contentionem neque per inanem gloriam, sed 

in humilitate alter alterutruni maiorem aestimantes: cf in Rom. 

xii 10; in 1 Cor. xiiii 40; xvi 14; in Gal. v 15; in 1 Thess. ii 6. 

2 Tim. ii 11 si conmortui sumus, et coniiiuemus: in 1 Cor. i 9; 
cf in Rom. vii 4; in 2 Cor. iiii 10; in Phil, iii 11; in Col. iii 1. 

Twelve verses are quoted or alluded to four times: — 

Gen. i 26 faciamus hominem ad imaginem et similitudineni 
nostram, etc.: cf in Rom. viii 21; in Eph. iiii 24; in Phil, ii 6; in 
1 Tim. ii 9. 

Math. V 39 siquis te percusserit in dextera maxilla tua, praehe 
illi et alteram (aliam): cf in Rom. xii 17 ; in 1 Cor. ii 14; iii 18 ; vi 8. 

Math, xxiiii 13 qui perseuerauerit usque in Jinem, hie saluus 
erit: in Rom. ii 7; in Gal. vi 9; in Phil, ii 13; in 1 Tim. vi 14 
(cf also in 1 Thess. ii 12). 

Act. xiii 2 segregate mihi Barnahan et Saulum ad opus quod 
elegi eos: in Rom. i 1 ; cf in Rom. i 5; in 1 Cor. vi 2; in Tit. i 3. 

Act. XX 28 uidete gregem in quo uos spiritus sanctus episcopos 
ordinauit, regere ecclesiam dei, quam adquisiuit sanguine suo: cf 
in Eph. i 14; in Phil, i 2; in 2 Thess. i 3; in 1 Tim. iii 8. 

Rom. iii 20 ex operibus legis non iustificatur omnis caro coram 
illo: per legem enim cognitio peccati: cf in Rom. iii 21 ; x 2, 3; in 
Gal. ii 14. 

1 Also quoted in c. 19 of Pelagius's Epistula ad Demetriadem. 


Rom. vi 13 sed neque exhiheatis membra uestra arma iniquitatis 
peccato, sed exhihete uos deo, tamquam ex mortuis uiuentes, et mem- 
bra uestra arma iustitiae deo : cf. Rom. xii 1 ^ ; x v 24, 29 ; in 1 Cor. x 33. 

1 Cor. V 1 omnino auditur inter uos fornicatio, et talis fornicatio 
qualis nee inter gentes, ita ut uxorem patris aliquis habeat: arg. 
omn. epist. ; in 2 Cor. ii 6; vii 12; xii 21. 

1 Cor. xii 11 haec autem omnia operatur iinus atque idem 
spiritus, diuidens singidis proiit uidt: in 1 Cor, xii G; cf. in Rom. 
xii 3; in 2 Cor. iii 18; viiii 14. 

1 Cor. xiii 7 (caritas) omnia suffert, omnia credit, omnia sperat, 
omnia sustinet: cf. in 1 Cor. iiii 10; in Phil, i 7; in 1 Thess. i 3; 
in Col. iii 14. 

Eph. V 30 membra sumus corporis eius, de came eius et de ossi- 
bus eius: in Rom. vi 6; in 1 Cor. xii 27; cf, in 1 Cor. vi 15; in Gal. 
iii 27. 

Phil, ii 8 humiliauit semet ipsuvi, /actus oboediens usque ad 
mortem, mortem autem crucis: in Rom. i 1; cf, in Phil, ii 5, 12, 22. 

About fifty verses are quoted or alluded to three times: — 

Gen. iii 4, 5 dixit autem serpens ad midierem: Nequaquam 
morte moriemini: scit enim dens quod in quocumque die comedentis 
ex eo, aperientur oculi uestri et eritis sicut dii scieiites bonum et 
malum: cf. in 2 Cor. xi 3; in Gal. vi 7; in 1 Tim. vi 21, 

Ps. xiii 1 dixit insipiens in corde sua: Non est deus...non est 
qui faciat bonum, non est usque ad unum: in Rom. iii 10; v 12; 
cf. Eph. iiii 18. 

lob i 21 7iudus egressus sum de utero matris meae et nudus 
reuertar illuc.sit nomen domini benedictum: in Eph, v 20; cf, in 
1 Thess. V 18; 1 Tim. vi 7, 

Esai. Iiii 4 ipse peccata nostra portauit: in Rom. vi 10; in 
1 Cor, XV 3; in Col. iii 13. 

Math, vii 23 numquam noui uos; discedite a me, aperarii ini- 
quitatis: cf. in 1 Cor. xiii 2; xiiii 38; in Gal. iiii 9. 

Math. X 10 dignus est enim operarius cibo suo: in 2 Cor. viii 15 ; 
cf, in 1 Cor, viiii 4; in 2 Cor. xi 7. 

Math, xiii 43 tunc iusti sicut sol fulgebunt: in 1 Cor. ii 9 ; cf. in 
Rom. ii 7 ; in 2 Thess. i 10. 

1 This passage may however be taken to refer to a later part of the same verse. 


Math, xiii 52 omnis scriba doctus in regno caelorum similis est 
homini patri familias qui profert de thesauro suo noua et uetera: 
in 1 Cor. i 6; cf. in 2 Cor. v 18; in 1 Tim. vi 17. 

Math, xviiii 21 uade uende omnia quae habes et da pauperibus, 
et habebis thesaurum in caelo, et ueni sequere me: in 1 Cor. xiii 3; 
in 1 Tim. vi 19; cf. in 1 Tim. vi 17. 

Math. XXV 41 discedite a me, maledicti, in ignem aeternuin, qui 
praeparatus est diabolo et angelis eius: in 2 Cor. v 10; in Gal. iii 
10; cf. in Eph. v 5. 

Marc, vii 8 traditioneni hominum: cf in Gal. ill; in Tit. i 10, 14. 

Luc. X 7 dignus est operarius mercede sua: in 1 Cor. viiii 14; 
in 2 Thess. iii 9; cf 1 Cor. viiii 4 (cf Math, x 10 above). 

Liic. xi 46 nobis legis peritis uae: quia oneratis homines on ribus, 
etc.: cf in Gal. iii 5; v 7; in Eph. ii 14. 

Luc. xxiii Z^ pater, dimitte illis: non enim sciunt quid faciunt: 
cf in Rom. viii 9; in Phil, iiii 7; in Col. iii 15. 

loh. i 3 omnia, per ipsum facta sunt: cf in Rom. xi 36; in 1 Cor. 
i 9; viii 6. 

loh. iii 5 nisi quis renatus fuerit ex aqua et spiritu, non potest 
introire in regnum dei: cf in Gal. iiii 19; vi 15; Eph. ii 10. 

loh. vi 56 qui manducat corpus meum et bibit meum sanguinem, 
in me manet et ego in eo: in 1 Cor. xi 25; cf in Rom. viiii 2. 

Act. iiii 32 multitudinis credentium erat cor unurn et anima 
una, nee quisquam suum dicebat, sed erant illis omnia communia : 
in Eph. iiii 3; in Phil, i 27; cf in 2 Cor. viii 15. 

Act. V 5 audiens Ananias haec uerba cecidit et expirauit: cf in 
Rom. xiii 4; in 1 Cor. iiii 21; in 2 Cor. xiii 9. 

Act. xiii 11 eris caecus, Qic... confestim in eum cecidit caligo, etc.: 
cf in Rom. xiii 4; in 1 Cor. iiii 21; in 2 Cor. x 4. 

Act. XV 10 nunc ergo quid temptatis deum, inponere iugum 
super ceruicem discipulorum, quod neque patres nostr'i neque nos 
portai'e potuimus? cf in Gal. ii 4; v 1; in Tit. i 15. 

Rom. vi 6 uetus homo: cf in Rom. vi 4; xiii 14; in Gal. ii 20. 

Rom. vii 12 lex quidem sancta et mandatum iustum et sanctum 
et bonum: in Rom. v 7; cf. in 2 Cor. iii 7; cf 1 Tim. i 8. 

Rom. viii 24 spe salui facti sunms: spes autem quae uide- 
tur, non est spes: in Rom. xii 12; in 2 Cor. iii 12; cf in Phil, 
iii 12. 


Rom. viiii 1 testimonium mihi perkibente conscientia niea in 
spiritu sancto: cf. in 2 Cor. iiii 2; xii 17; in 1 Tim. i 5. 

Rom. xiii 10 dilectio proximi malum non operatur: plenitudo 
ergo legis est caritas (dilectio): in Rom. iii 28; in 1 Cor. xiii 2; cf. 
1 Cor. xiii 2. 

1 Cor. vi 19 membra uestra templum est spiritus sancti: cf. in 
Rom. viii 11; 1 Cor. iii 16; 2 Cor. v .5. 

1 Cor. x 33 sicut et ego omnibus per omnia placeo, non quaerens 
quod mihi utile est, sed quod m^dtis, ut salui fiant: in Rom. xv 2; 
in 1 Cor. viiii 22; cf in 1 Cor. xiii 5. 

1 Cor. xii 28 et quosdam quidem constituit deus in ecclesia... 
deinde uirtutes, exin gratias curationum: cf in Rom. xii 3; in 
1 Cor. iii 5; in Eph. iiii 11. 

1 Cor. xiii 5 {caritas) non quaerit quae sua sunt: in 1 Cor. viii 1; 
cf in 1 Cor. ii 14; in 2 Cor. iiii 5. 

1 Cor. xiiii 25 occulta etiam cordis eius manifesta fiunt, et tunc 
cadens in faciem adorabit deum, pronuntians quod deus uere est in 
uobis: in Eph. v 13; cf in 2 Cor. i 23; in 1 Thess. ii 4. 

Gal. i 12 neque enim ego ab homine accepi illud neque didici, 
sed per reuelationem lesu Christi: cf in 1 Cor. xi 23; xv 3; in 
Gal. i 13. 

Gal. iiii 7 itaque iam non est seruus, sed filius: quod si filius, et 
heres per deum: cf in Gal. iiii 21, 24, 29. 

Eph. iiii 22 deponere uos secundum pristinam conuersationem 
ueterem hominem, qui corrunipitur secundum desideria erroHs: cf 
in 1 Cor. v 7; xv 49; in Gal. ii 20. 

Eph. V 22 midieres uiris suis subditae sint, sicut domino: cf in 
1 Cor. xi3, 4; in 1 Tim. v 13. 

Eph. V 27 ut exhiberet ipse sibi gloriosam ecclesiam, non ha- 
bentem macidam aut rugam aut aliquid eiusmodi, sed ut sit 
sancta et inmaculata: cf in 1 Cor. i 2; in 1 Tim. iii 2; in 2 Tim. 
ii 20. 

Phil, i 1 Paulus et Timotheus, serui... omnibus Sanctis... cum 
episcopis et diaconis: in 2 Cor. i 1; cf in 2 Cor. i 1; in 1 Tim. 
iii 8. 

Col. i 24 corpore eius quod est ecclesia: cf in 1 Cor. xv 28; in 
Eph. ii 21; iiii 12. 

Col. ii 5 et si corpore absens sum, sed spirit^i uobiscum sum. 


fiaudens et videns ordinem nestrum: arg, omn. epist.; in 1 Cor. ill; 
in 1 Thess. ii 17. 

1 Tim. V 23 noli adhiic aquam hibere, sed uino modico utere 
propter stomachum tuum et frequentes tuas injirmitates: cf. in 
1 Cor. xi 31; in 1 Tim. iii 8; in Tit. ii 3. 

2 Tim, iiii 7 honum eertamen certaui, cursum consuTnmaui, 
fidem seruaui: in Rom. viiii 16; cf. in Gal. v 7; arg. in 2 Tim. 

Tit. i 5 ut...constituas per ciuitates preshyteros: cf. in 1 Cor. 
iii 5; arg. in Tit.; in Tit. i 7. 

Hebr. i 3 imago expressa suhstantiae eius: in 2 Cor. iiii 4; cf. 
in Phil, ii 6; in Col. i 15. 

Hebr. x 34 nam et uinctis conpassi estis et rapinam honorum 
uestrorum cum gaudio suscepistis, cognoscentes uos habere meliorem 
et manentem substantiam: arg. omn. epist.; in 1 Thess. ii 14; cf. in 
Gal. ii 10 (cf pp. 70 f). 

1 loh. iii 6 omnis qui in eo nianet, non peccat; et omnis qui 
peccat, non uidit eum nee cognouit eum: in Rom. iii 11; in Eph. iiii 
13; cf in Eph. i 17. 

Apoc. vi 8 qui sedebat desuper, nomen illi peccatum^ et mors, et 
inferus sequebatur eum: cf in Rom. vii 8; in 1 Cor. xv 26; in Col. 
iiii 11. 

I venture to think that this is a considerable body of evidence 
in favour of common authorship for all the expositions. The list 
may have at the same time some interest as a collection of the 
author's favourite verses of scripture, shedding further light on his 
personality and point of view. 

(e) Community of Style and Language 

The present section has not the slightest claim to be con- 
sidered a complete account of the style and language of the 
commentaries, or even a complete account of the most character- 
istic elements in these. I doubt not that another student might 
select many other instances of equal cogency with those adduced 
below. It happens that Pelagius is a very correct writer, with 
little of the extraordinary about his vocabulary or idiom, in this 
respect differing very greatly from his predecessor, Ambrosiaster. 
' On this reading see below, chap, iv p. 173. 


The collection of characteristic expressions has, therefore, cost 
much more trouble. 

The evidence in this case consists rather of many examples of 
a few simple phrases than of few . instances of an unusual type. 
The portion concerned with Grammar is particularly brief 

I. Grammar 

1. Accidence 

The ablative of uetus is generally, perhaps always, uetere: Rom. 
vii 12; 1 Cor. vii 5; viiii 20; Gal. v 23; Eph. iiii 24; Tit. iii 6. 
(Ambst always, Hier. usually ueteri.) In abl. sing, of compar. 
adj. we sometimes find -e, e.g. meliore (1 Cor. iii 12), and perhaps 
this form should always be read. 

The geniti»e plural of present participles ends in -um pro- 
bably more frequently than -ium: examples of -um are 1 Thess. 
ii 18; 2 Tim. ii 4; 2 Tim. iiii 5. 

In the matter of word formation, several instances where the 
preposition ex is concerned, may be mentioned. Just as pro consule 
is an earlier stage than the inflected substantive proconsul, so ex 
praefecto precedes expraefectus in time. At what date the latter 
type begins, it is not perhaps possible to state exactly, but Pelagius 
knows nothing of it, for in him we find only the earlier stage: 
Rom. xvi 23 hie arcariam ex arcario dicit, stent gentes credentes 
ex gentibus saepe nominauit; Eph. arg. Ephesii ex ludaeis et 
gentibus; Phil, iiii 3 Clemens ex philosopho, magnae doctrinae 
uir, qui Romae episcopus fiiit; Col. iiii 14 (Lucas) ex medico erat, 
sicut Matheus iam apostolus adhuc dicitur puhlicanus; 1 Tim. arg. 
rationem reddens quod non sit minim, si ipse ex persecutore 
saluatus sit. 

2. Syntax 

The author shows an extreme fondness for the ablative of the 
gerund, not as a mere substitute for the present participle, but 
with the earlier, instrumental force. The following list of examples 
is fairly complete : — 

Rom. ii 4 bonus est expectando, iustus est puniendo; 12, 15, 27 ; 
iii 4, 28; iiii 15; v 10; vii 13; viii 3, 6; viiii 17, 22; xi 4 bis: 
xii 5, 19 bis, 21; xiiii 8 bis, 17, 23 bis; xvi 4. 


1 Cor. i 19 elegendo piscato7'es, artem rhetoricam et philosophi- 
cain reprohauit...ut...paulatwi meditando prq/icia^; 28; vi 19, 20; 
vii 8, 34 his; viii 8 his; viiii 13, 27; x 5; xv 51. 

2 Cor. ii 11 ut peccatores pereant desperando, quo modo in 
nimia remissione minime corrigendo ; iii 1, 9 ; vii 3 ; viii 14; x 14; xii 7. 

Gal. iiii 24 reprehendendo atque uindicando ostendit ilia non se 
propter sua mysteria Jier^i uoluisse; v 14, 15. 

Eph. ii 3 in qiiibus et nos ovines conuersati sumus aliquando]. 
Non solum uos gentes non credendo, sed et nos ludaei peccando; 
5, 13; iii 8; iiii 8, 16; v 3, 6; vi 4. 

Phil, i 5 quod ah initio communicatis praedicando; 6, 16, 20; 
ii 2, 5 ter, 7, IS ter, 27; iii 13. 

1 Thess. arg, laudando illos apostolus ad maiora prouocat; 
i 10; iiii 4. 

2 Thess. iii 5 dirigat]. Reuelando quanta sint quae . . .repro- 

Col. i 10 quo modo deus det uelle et adiuuet uel confirmet, do- 
cendo scilicet sapientiam et intellectvs gratiam tribuendo, non 
lihertatem arhitrii auferendo; ii 13, 15 quinquiens. 

1 Tim. i 19 habens honam conscientianij. Implendo quod 
daces... male uiuendo ipsam quoque fidem perdiderunt; ii 8; v 3 bis; 
vi 2, 4 bis, 13, 17. 

2 Tim. ii 10 cum has passiones potuerim iam raoriendo finire; 
iii 12. 

Tit. i 7 turpe lucrum adulando sectari; ii 6. 

Another favourite construction is that of the present participle 
with a noun, particularly an abstract noun in the singular, in the 
ablative absolute. Examples are : — 

Rom. ii 25 carnis circumcisione cessante ; v 4, 14; vi 15; 
vii 9; viii 4; viiii 17; xi 24; xiii 7, 11, 13. 

1 Cor. vii 5 uno se retrahente; viiii 10; xiii 3; xiiii 14, 22. 

2 Cor. iii 3 confirmante per uirtutem spiritu sancto; 13, 18; 
V 1, 4, 19; vii 4. 

Gal. iiii 24 manente historiae ueritate; v 16 his. 
Eph. arg. Paulo apostolo praedicante; ii 1, 5; v 22. 
Phil. arg. ipso praedicante. 

Col. ii 16 {umbra) cessauit corpore ueniente, quia imagine opus 
non est, ueritate praesente; 23. 

s. p. 6 


2 Tim. ii 19 deo illos saos esse optime cognoscente. 

Tit. ii 12 perfede, domino ipso dicente — 

Philem. 8 caritate faciente obsecj'ure malumus quam iubere\ 

It may be here noted that Pelagius is throughout very strict 
in regard to the sequence of tenses. Examples need not be ad- 

Object sentences in the later authors are especially interesting. 
In addition to the accusative and infinitive construction, we fre- 
quently find examples of a native Latin colloquial construction 
with quod, and also, on the analogy of on with its double sense, 
a Biblical Grecism, with quia and quoniam. Even ut is found. 
The later authors can be classified according to their practice in 
this matter. The few severe purists show the accusative and 
infinitive only ; the somewhat less strict show also the use of quod; 
the still less strict add quia, and the really lax indulge also in 
quoniam: ut in this use is absent from most writers. I have printed 
statistics on this matter elsewhere^ and will not repeat them here. 
Quoniam and perhaps tit are entirely absent from Pelagius, and 
quod is commoner in him than quia. These facts at once place 
him in the better class of writers. If my numeration is correct, 
the proportion of cases is about 98 of quod to 72 of quia. A possible 
example of ut occurs at Rom. v 15. To avoid repetition, only some 
of the examples of the quod and quia constructions will be given 
at this stage: others will be found later in the chapter ^ Quud 
and quia are followed sometimes by the indicative, sometimes by 
the subjunctive, but while the subjunctive is much more frequent 
with quod, the indicative predominates with quia. 

quod: (indie.) Rom. arg. unde prohatur quod seruiuimus; viiii 
11; X 3, 5; 2 Cor. viii 3. 

(subj.) Rom. i 2 didt quod... sit promissum...e^ quod... sit crea- 
tus; 19 bis; iii 2bis; v 1; vii 8; viii 13, n, 33-4; viiii 4, (14), 33: 
x5,14; xiiiil8; lCor.il9; vii 3; viiii 20, 21; xii6; xv 28; 2 Cor. 
iii 6 bis; v 11; viii 15; xi 16, 33 bis; xii 6, 9; Gal. i 12; ii 11; 
iii 11; iiii 28; v 9; Eph. i 1; ii 2; iii 1, 18; Phil, ii 5; 1 Tim. iii 11. 

1 See also under ezisto below. 

2 In Phil, i 7 is an exception. 

'■^ In my linguistic introduction to Sanday's Novum Tex (amentum S. Irenaei 
(Old-Latin Biblical Texts, no. vii). 

* Under notare, adtendere, considerare etc. on pp. 85, 89 f. etc. 




quia: (indie.) Rom. iii 4 6is; v6, 15; viiii 16, 30 6is, 33; x2 6is; 
xi 10, 18; XV 16; 1 Cor. i 4, 23, 26; v 6; xi 4 his, 26 his\ xv 6, 11, 
28; 2 Cor. v 5; vi 11; xi 12; Gal. iiii 29; Phil, i 27; iii 13; iiii 15; 
Col. iiii 18; 2 Tim. i 18; iii 1; iiii 20. 

(subj.) Rom. i 19; v 1; vii 25; viii 8; xi 6; xiiii 2, 11; 1 Cor. 
vi,i 14; viii 7; xi 27; 2 Cor. ii 3; viii 22; Gal. iiii 29; Philem. 22. 

At least two points in regard to the arrangement of words call 
for remarks First, Pelagius is very fond of separating a noun, 
usually the object, from its adjunct by interposing the governing 
verb. The adjunct may be an adjective or participle in agreement 
with the noun, or it may be another noun or pronoun in the genitive 
case governed by the noun. Sometimes the adjunct consists of both 
adjective and dependent genitive. Examples occur in abundance 
in the Argumentum Omnium Epistularum and right through to 
the end of the work. It is not necessary to mention more than a 
few: — 

Arg, omn. epist. praesentia atque orientia resecaret uitia. 

„ „ „ futuras excluderet quaestiones. 

„ „ „ rediuiua semper populi conpressere peccata. 
Rom. i 8 idolorum nimia fuerant cultura possessi. 

iiii 24 eius quasi patris imitemur exemplum. 

1 Cor. iii 19 suam non reseruat iniuriam. 

vii 3 proprias in matrimonium, acceperunt sorores. 

2 Cor. ii 6 paternum j)oUuit tormn. 

V 16 nullius ueterum im^itamur exemplum. 

Gal. iiii 16 omneiii continentiam suo illis monstrahat exemplo. 

Eph. ii 12 unum amiserunt uerum. 

Phil, iiii 9 ilia explanant superioi'a. 

1 Thess. i 5 iustitiae covim,endatur exemplo. 

2 Thess. ii 12 homini crediderunt diabolica arte falleiiti. 
Col. iii 2-3 caelestem sectamini sapientiam. 

1 Tim. i 9 occasiones fugere delictorum. 

2 Tim. i 16 ad exemplum prouocet bonum. 
Tit. ii 1 ad uitam pertinent Ghristi. 
Philem. 8 ad rem pietatis pertinet Ghristianae. 

The second point is the matter of rhythm. There can be no 

Dr Alfred J. Smith first called my attention to these points. 



doubt that Pelagius, in common with other careful writers of the 
imperial period, paid attention to the rhythm of his sentences. It 
was not necessary for my purpose to make a list of the different 
types of clausula employed by him, or to calculate the number of 
times each occurs. It is enough to call attention to the frequency 
with which a sentence ends with four long syllables^ This ending 
is found in probably about a third of the total number of endings 
in the treatise. Another ending found frequently is the type 

- ^ . From the Argumentum Omn. Epist. we take sentences 

ending with trdnsmiserunt, sunt porrectae, ddquisUos, cmfllctarent, 
seruauerunt, constdntes Inuentl: the following occur later, to which 
many might be added: — Rom. i 3 non est fact fis; i 8 demonstrdtur; 
i 12 consolemftr ; 1 Cor. i 5 testdmenti; i 25 sunt sdludil; 2 Cor. 
ii 5 'profeclstis; contrlstduit; Gal. ii 17 Indulgere; Eph. i 3 infra 
scriptd; Phil. arg. PJdlippenses in Christo; 1 Thess. iiii 7 non 
permUtit; 2 Thess. iii 1 exor^erjius; Col. iiii 3 in doctrind; 1 Tim, 
vi 2 praemium ostendendo; 2 Tim. ii 2 confirmdui; Tit. i 16 quae 

quaeruntur; Philem. arg. correctUrUm. The ending -^ , 

being one of Cicero's favourites, usually numbered 1 by the 
students of metrical prose^, need not be illustrated. 

The only figure to which I shall call attention is that of ellipsis. 
The instances are not numerous, but are sufficiently interesting to 
have a value as evidence: Rom. xi 2 in Regnorum, ubi scriptum 
est de Helta; 1 Cor. vii 5 unde et in Regnorum panem sacerdotalem 
non nisi continentes accipiunt; the omission of lihris in such phrases 
is quite in line with the practice of St Cyprian, who wrote In 
Basilion simply ^ Another example of ellipsis occurs at Rom. vii 12 
deus non numquam in Vetere 'bonus' et in Nouo dicitur 'iustus,' 
where testamento is understood, and at Rom. viiii 4 Veteris latio et 
Noui promissio. 

Occasionally Ave find what may be called philological notes: in 
1 Cor. XV 25 'donee' no7i semper finent significat, sicut est illud: 

^ The spondaic cadence is rare in Cicero, but frequent in Livy : see A. C. Clark, 
The Cursus in Mediaeval and Vulgar Latin (Oxford, 1910) pp. 7 f. 

2 Of. Clark, p. 7. 

3 Cf. C. H. Turner in Journal of Theological SUidies vol. vi (1904—5) pp. 249, 
259 f., 268. Max.-Taur. c. lud. eh. 9 1. 8 (ed. C. H. Turner, J.T.S. vol. xx (1918—19) 
]). 306) also has in Regnomm : so also Ps.-Aug. Sjjecidum. 


ego deus uester donee senescatis, et cetera talia^; in 1 Cor. 
XV 31 'per' non semper significatio iurarnenti est: nam si dicavi 
'per' puerimi misi, non statim per piierum iurasse recte putahor; in 
Gal. iii 4 hie 'si tamen' non dubitantis sermo sit, sed potius con- 
Jii^mantis, secundum illud: si tamen iustum est aput deum 
retribuere his qui uos tribulant, tribulationem. 

II. Lexicography. 
Details of Phraseology and Vocabidary 

(a) Favourite Openings of Notes. 

One of the most constant features of the commentary is the 
phraseology at the beginning of the notes. Without attempting 
to exhaust the formulae employed for this purpose, we can easily 
adduce a sufficient number of examples of each to show unity of 
authorship throughout the commentary. A comparison with the 
methods of Ambrosiaster shows at once the difference of author- 
ship here-. 

Notandum quod^: (a) c. indie. 1 Cor. vii 9; xv 45; 2 Cor. viii 7; 
xiii 11; Gal. v 21; Phil, iiii 18; 1 Thess. iiii 18; Tit. i 6. 

(b) c. subjunct. Rom. v 8; xii 3; 1 Cor. x 8; xii 12; xiii 3; 
2 Cor. vi 6; Eph. ii 2; 1 Thess. i 5; Col. i 29; 1 Tim. ii 15; 
Philem. 14. 

Notandum quia: c. indie. Rom. viii 2; 2 Cor. xi 23; Gal. i 22; 
Eph. i 17; iiii 18; Phil, i 10; 1 Thess. v 11: no examples c. sub- 

Notandum c. ace. et inf Rom. xv 23; 2 Cor. vi 5; 1 Tim. v 7. 

Totus: Rom. xii 9, Tota puritas debet esse in Ghristiano; 1 Cor. 
XV 2 Tota ratio praedicationis nostrae haec est; 1 Cor. xv 41 Tota 
comparationis huius diuersitas ad hoc facit; 2 Cor. viiii 2 Tota 
prouincia, cuius caput estis 

Recapitulo^: Rom. vii 25 Recapitulat, ut concludat; Rom. xiii 9 
Recapitulatur omnis iustitia in proximi dilectione. 

1 This note was borrowed from .Jerome, see chap, v p. 184. 

2 Cf. Study of Ambrosiaster pp. 64 f. 

3 Cf. Simul notandum quod, p. 90, etc. 
■• The word also Rom. viiii 30. 


Quaeritur^ etc. Argum. omn. epist. pi'imum quaeritur qua re; 
Rom. i 1 Quaerimus qua re; 2 Cor. i 1 Quaeritur cur; 1 Cor. i 11, 
1 Thess. iii 2 Quaeritur quo modo. 

Vult ostendere- Rom. iii 26; xi 13; 2 Cor. iii 5='; Phil, ii 5 
(followed by ace. et inf.); Rom. v 6 (followed by quia c. indie), 
viii 31 (followed by quod c. subjunct.); cf Rom. v 11 Hie ostendere 
uult (followed by ace. et inf ); 2 Cor. vii 5 ostendere. . .uult {quantam 
c. subjunct.). Vult is the first word of Rom. vi 2; x 18; 1 Cor. iii 4; 
Gal. i 11; 1 Tim. v 18: cf 1 Cor. xii 31. 

Hie ostendit^ Rom. i 10, 11; viiii 9; x 1; 1 Cor. iiii 20 (osten- 
ditur); xiiii 22 (ostenditur) ; 2 Cor. xii 6 (uidetur ostendere); 
Phiiem. 22 (ostenditur). 

Ostendit Rom. i 2; ii 14; vi 4; xv 14, 21; xvi 5; 1 Cor. i 2; 

V 1; viiii 20; xv 34; 2 Cor. viii 10; xi 6; Gal. i 1 (ostenditur); 
i 4, 18; iiii 15; Phil, i 4; 1 Thess. iii 10; 1 Tim. iii 13; cf 1 Tim. 

V 23. 

Contra and Hoc contra-^ etc.: Rom. xi 22 Contra eos 
contra eos qui,..; 2 Cor. iiii 6 Hoc contra omnes inimicos...; Gal. 
i 19 Contra eos, qui...; Col. i 16 Contra Manicheos; Gal. v 24; Hoc 
contra illos qui ; 2 Tim. ii 8 Hoc contra illos, qui; Rom. viiii 2 
Contra ludaeos acturus; xv 25 hoc contra illos facit; 1 Cor. i 26 
Hoc contra illos agit; viiii 21 Contra Arrianoset Fotinianos; xiii3 
contra eos qui. ..; xv 45 quod contra Manicheos et Apollinaristas 
facit; Eph. v 5 Contra illos agit qui; Col. ii 8 Contra philosophos agit. 
Compare also the passages with Hie locus, etc.; Rom. i 2 totus hie 
locus contra Manicheos facit; xi 36 simul etiam contra A7Tianos 
facit hie locus; 1 Cor. i 9 hie locus contra Arrianos facit; cf. 
Gal. i 8 facit autem sententia haec contra omnes haereticos. 

Ne frequently introduces a note, e.g. 1 Cor. vii 35 Ne necessi- 

1 Similarly in body of notes; with quare Eom. ii 26, 2 Cor. viiii 11; with ctir, 
1 Tim. iii 8; with quo modo Rom. iii 19; iiii 18; 1 Cor. vii 16; xiii 3; xiiii 34; 
with si Gal. iii 10; with unde 1 Cor. i 19; with ut Rom. viii 10; with quid 1 Cor. 
vii 3 bis. 

- In Origen-Rufinus in Eom. (see Dr Smith in Journ. Theol. Studies vol. xx 
(1918—19) p. 142 n.). 

3 Also 2 Cor. V 10; Gal. i 10, 13; though not the first words. 

■* This word predominates over monstro : examples in the body of the notes are 
Rom. iii 19, 29; viii 3; xv 25; 1 Cor. i 26; ii 1 ; xi 4 ; 2 Cor. xi 1, 33; Gal. ii 11; 
iiii 19 ; v 9 ; Eph. ii 10. 

5 Also in body of note, 2 Thess. i 8. 


tatem...; 1 Cor. x 1 Nequis; Gal. i 19 'Ne...iiideretur (cf. Rom. x 3; 
xiii 6); but is found in especial frequency with parts of dico and 
puto: ne dicerent Rom. iii 19; iiii 11; v 20; vi 2, 15; xi 6; 1 Cor. 
XV 33; 1 Tim. i 11; nequis diceret Rom. i 82'; l^equis forte di- 
ceret Rom. ii 14; Gal. v 9; 'Nequis dicat 2 Cor. viii 20; Ne diceret 
2 Tim. ii 6; cf Ne...adsererent Rom. xv 12. Nequis putet- 1 Cor. 
X 1, 29; Phil, iii 4; (1 Tim. iii 6). Non putet 1 Cor. vii 18. Ne 
putetis^ 1 Cor. ii 6; xiiii 18; 2 Cor. vii 12; viii 12; xi 5; Eph. iii 13; 
cf ut putetis Gal. iiii 12. Ne putaret* 1 Tim. v 23. Ne...puta- 
remus Rom. xii 141 Ne putarent 1 Cor. xv 8; Gal. ii 2^ Ne 
putaretur'' Rom. v 15; Gal. i 2. 

Expono®: Rom. ii 23 Exposuit quid sacrilegium dixerit; Rom. 
iiii 4 Exponit exemplum; Rom. xiiii 21 Exponit quid sit per 
quod...; Rom. xv 22 Exposuit illad quod in capite dixerat; 1 Cor. 
xi 15 Exposuit quod dixerat; 1 Cor. xiii 11 Exposuit qualiter 
euacuentur ilia...', 1 Cor. xiiii 3 Exposuit qiiot modis intellegenda 
sit proplietia; 2 Cor. iii 9 Exposuit quod superius dixerat; Eph. 
iiii 9 Exponit cur dicatur ascendisse; Eph. iiii 22 Exposuit quid 
sit uetus homo; Eph. iiii 24 Exposuit quid sit hominem ad imagi- 
nem. dei esse creatum; Eph. v 18 Exponit continentiae utilitatem; 
Col. iii 10 Exposuit imaginem in actu consistere. 

Repeto": Rom. xi 15 Repetit quod dixerat superius; Rom. xiiii 
20 Repetit quod superius dixerat; 1 Cor. viii 4 Repetit quod su- 
perius. . .dixerat; 2 Cor. xii 4 Quod iterum repetit, ostendit...; Gal. 
i 9 Repetitum fortius commsndatur; Phil, iiii 4 Repetit ut magis 
ac magis gaudium confirmetur: cf Tit. iii 14 Id ipswm repetit, 
quod superius dixerat. 

Incipio^": Rom. i 18 Incipit ad partem gentiuni loqui; 1 Cor. 

^ Also in body of note, 1 Cor. xi 16 ; cf. ne de nobis dicatur, Col. iii 4. 

■■' Also in subordinate clauses, Rom. viiii 14; 2 Cor. x 10; cf. 1 Cor. xii 28. 

^ Also in subordinate clauses, 1 Cor. xiiii 33; 2 Thess. iii 13; cf. 1 Cor. iii 1; 
2 Cor. i 1; 2 Cor. viii 1; Gal. iii 20. 

^ Also in subordinate clause, 1 Cor. i 16. ^ ne piUnnux 1 Cor. xii 28. 

® Also in subordinate clause, 2 Cor. i 1. 

^ Also in subordinate clauses, Rom. v 2; 1 Cor. v 3; xv 6; 2 Cor. i 17; xi 33 ; 
Gal. i 14 ; Tit. i 8. 

8 Also in body of notes, Rom. vi 12 ; viii 13, 19 ; 1 Cor. i 9 ; xiiii 17 ; Gal. vi 17 ; 
Eph. arg.; Col. i 10; 1 Tim. iii 5. 

'^ Also in body of notes, Phil, iii 1. 

1'^ See also under hinc (p. 88), causa (pp. 95 f.) ; and cf. 1 Thess. iiii 1. 


vii 1 Incipit de coniugiis; 1 Cor. vii 8 Incipit aliam causam, de 
immptis; 1 Cor. viii 1 Incipit de idolothytis; 1 Cor. xi 2 Incipit 
dfi uelamentis; 1 Cor. xi 18 Incipit de sacramentis; Eph. ii 1 Incipit 
collata heiieficia replicare; Eph. v 13 Incipit lumen esse cum credi- 
derit; cf. also 1 Thess. iiii 1 Post laiidem et consolationem mcvpit 

Hinc'; lam hinc: Rom. ii 17 Hinc conuertitur, etc.; Rom. vii 1 
Hinc incipit difficultatem legis ostendere; Rom. vii 7 Hinc in per- 
sona eius hominis loquitur; Rom. viii 8 Biinc probata r quia superius; 
Rom. viii 18 Hinc uult futuram gloriam commendare; Rom. x 17 
Hinc 7-esponsio apostoli; Rom. xiiii 1 Hinc oblique illos increpare 
incipit; Rom. xiiii 2 Wino, prohatur quia non; Rom. xiiii 13 Hinc 
subtiliter ingreditur; 1 Cor. i 10 Hinc iam causam contra dis- 
sensionem adgreditur; 1 Cor. iiii 7 Hinc ad ipsos inflatos per 
eloquentiam uerba conuertit; 1 Cor. iiii 18 Hinc incipit causam 
fornicationis arguere; 1 Cor, vi 1 Hinc probatur ChHstianos 
tunc...; 1 Cor. vii 39 Hinc incipit de digamis et uiduis; 2 Cor. 
iii 3 Hinc ue7^e cognoscimini; 2 Cor. iii 3 Hinc iam differentiani 
inducit; 2 Cor. iiii 7 Iam hinc incipit ostendere; 2 Cor. xii 9 Hinc 
intellegimus etiam sanctos; Eph. iiii 25 Hinc describit ipsas species 
sanctitatis; Phil, i 12 Hinc consolatur eos; 1 Tim. ii 4 Hinc ^ro- 
batur deum nemini...; 2 Tim. iiii 20 Hinc probatur quia 

Hie (pron.), in various parts of the word, very frequently, but 
especially in the phrase Hoc est, e.g. Rom. i 9 Hoc est: in toto corde, 
etc.; Rom. i 15 Hoc est paratum; Rom. iii 8 Hoc fortassis ideo 
putabant; Rom. viii 12 Hoc tutum agit ut...; Rom. viiii 4 Hoc est 
ministerium angelorum uel prophetarum; Rom. xi 16 Hoc est, et 
primi et nouissimi, etc.; Rom. xiii 12; xv 4, 14, etc.; 1 Cor. iii 12 
Huius artis est qui...; 1 Cor. v 12 Hoc est de infidelibus; 1 Cor. 
vii 26 Hie 'existimo' non pro dubio posuit; 1 Cor. vii 29; viiii 12; 
2 Cor. vi 9 Hoc est, usque ad mortem peruenientes; Gal. ii 10 Hoc 
est, quod in omni... ; Eph. iiii 29 Hoc est quod alibi dicit; Phil, ii 11 
Hoc est, in natura et gloria deitatis; 1 Thess. iiii 10 Hoc est. 
etiam ignotos; 2 Thess. iii 10 Haec sit inquietudinis...emendatio; 

1 Tim. ii 9 Haec sunt ornamenta feminae Christianae; 2 Tim. ii 8 

1 In body of note, 1 Cor. vii 5 hina probatur quid possit perpetua continentia ; 

2 Cor. i 2 hinc iam gratias agunt deo; Eph. iii 21 hinc incipit moralia omni ecclesiae 
tradere instituta; 1 Tim. i 18 hinc dat auctoritatem ordinandi. 


Hoc contra illos, qui...; Tit. iii 8 Haec, non ilia quae sequimtur; 
Philem. 2 Hie diaconus erat^, etc. 

Hie locus^: 1 Cor. ii 8 Hoc loco... duae haereses abutuntur; 
1 Cor. xiiii 14 Oratio hoc loco oWs ratio. 

Hie (adv.)'': Rom. iii 4 Hie onines pro maxima parte dicit; 
Rom. V 11 Hie ostendere uidt; 1 Cor. viiii 24 Hie stadii cur sum 
iustitiae uel fidei comparauit; 1 Cor. xii 6 Hie uult ostendere; 

1 Cor. iiii 20; xiiii 22 Hie ostenditiir; Col. ii 11 Hie iam pseudo- 
apostolos taxat; Col. iii 16 Et hie ostenditur; Tit. ii 2 Hie senes... 
possunt intellegi. 

Reddit eausam (eausas): Rom. viiii 32 Reddit causas qua re 
non inuenerint iustitiam; 2 Cor. i 23 Reddit eausam non impleti 
promissi; 2 Cor. xi 12 Hie reddit eausam qua re non acciperet; 
Gal. ii 4 Reddit causas qua re circumciderit Titum; Eph. v 24 
Reddit iustam eausam subiectionis; 2 Thess. iii 11 Reddit superioris 
sententiae causas; 1 Tim. ii 14 Reddit eausas cur eas uelit esse 
suhiectas; 1 Tim, v 15 Reddit causas qua re eas nubere uelle prae- 

Reddit rationem (Reddita i^atione) with quare or quod is 
rarer: Rom. ii 13; iii 2; 1 Cor. xv 58*, 

Talis^: Rom. i 7 talis est ubique salutatio eius; Rom. x 4 Talis 
est qui Christum credidit...; Rom. xii 1 Talis ei placet hostia; Eph. 
v 9 Talem habet scientia fructum; 1 Thess. i 10 Tales estis ut...; 
Col. iiii 13 Tales erant primi temporis discipuli; 1 Tim. v 9 Tales 
uoluit eligi diaconissas; 2 Tim. i 2 Talis est salutatio Pauli ut; 

2 Tim. iiii 3 Tales sibi magistros inquirent; Tit. i 9 Talis est eli- 
gendus qui... ; Tit. iii 1 Talem te exhibe, ut — 

(b) Similar Phrases in the Body of the Notes, and Phrases 
introducing Biblical Quotations. 

adtendo: Rom. iiii 5 simul adtendendum quia...asseruit; 1 Cor. 

^ In body of note, cf. 1 Cor. iii 11 hoc solum agititr ne ; Gal. ii 11 hoc autein 
totum agit ut. 

2 Also in the body of the notes, e.g. Rom. iii 28 ; xi 3, 8; 1 Cor. xiiii 23 ; xv 24 
hoc loco; 1 Cor. iii 3 in hoc loco; Rom. i 2; viiii 17; xi 36; 1 Cor, i 9; xii 6; 
2 Cor, viiii 6 hie locus; Phil, ii 9 locus hie. See also under Contra above. 

■5 Cf. above under Vult ostendere, Hie ostendit. In body of note, Rom, i 23 hie... 
appellat; Rom. iii 28 hie... dicit; 2 Cor, ii 11 hie euidentissime ostenditur. 

4 It is commoner in the middle of a note : e.g. Arg. omn. epist. ; 1 Cor. xii 31 ; 
2 Cor. xi 16; Gal. ii 14; 1 Tim. arg, » So in body of note, Eph. vi 4. 


vii 35 simul adtendendum quia...erat; 1 Cor, xv 38 adtende quia 
...dixerit; 2 Cor. vi 10 simul adtendendum quod...dicit; Gal. vi 18 
simid et adtendendum quia...dixerit; Col. i 28 simul adtende 
quod...doceat; Philem. arg. nihil magis est in hac epistula adten- 
dendum nisi quanta 

simul notandum, or notandum: (a) quod c. indie: Eph. vi 13; 
Tit. iii 15. c. subjunct.: Rom. vi 13; xii 8; xiiii 17; 1 Cor. vi 1; 
viii 12; viiii 21; xiii 2; 2 Cor. xi 16; Gal. v 24; vi 2; Eph. iii 8. 

(b) quia c. indie: Rom. iii 24; Eph. iiii 29; 1 Tim, iiii 12. 

(c) c, ace, et infin.: Rom. viii 13; 1 Thess. iii 5; 1 Tim. v 8. 

(d) c. indir. interrog. : in quibus 2 Thess, iii 1; quo modo Col. 
i 23, 

caueo: Rom. i 31 caueamus ergo et nos ne...; Rom, v 9 cauea- 
mus er^o ne...; (Rom. xii 17 Tantum caue ne ideo facias...;) Rom. 
xiiii 8 cauendum ergo ne nos...; Tit. ii 14 caueamus ergo ne — 

considero; considerandum est: 1 Cor. xi 17 unde consider- 
andum est quid...; Eph. ii 4 unde considerandum est quantum; 
1 Thess, iiii 4 simul considerandum g'?aa...scribit. 

ne,.,uideatur (uideretur): Rom. xiii 6 ne uideretur Christus 
superbiam docuisse; Rom, xiiii 20 ne creaturam damnare uideatur; 
1 Cor. XV 27 ne deum. . .subiecisse uideretur; 2 Cor. v 5 ne ctii 
ivpossibile uideretur; 2 Cor. vi 18 ne cui inpossibile uideatur. 
But also at beginning of note, e.g. 1 Cor, xiiii 35 Ne uideretur eas 
etiam discere uetuisse; 1 Cor. xv 1 Ne illi hoc non credidisse uide- 

potest et ita (c, infin. pass.): Rom. vii 16 potest autem et ita 
intellegi; 1 Cor. viiii 20 potest et ita intellegi quod...dia;erit; 1 Cor. 
xi 12 potest et ita did (cf. 1 Cor. i 13 Potest ita did; xi 14 potest 
did); 2 Cor. vii 12 potest et ita did; cf. Tit. ii 2 senes et aetate et 
ordine possunt intellegi. 

quid (direct interrog.), in apodosis to si clause: 1 Cor. iii 2 Si 
ergo... illi... sunt..., quid de illis censendum est...l 1 Cor. iii 15 
quod si ille...erit..., quid de illis Jiet...'? 2 Cor. xi 17 si hie... 
negat..., quid de illis sentiendum est...'? 

quibus respondendum est: Argum. omn. epist.; Rom. viii 33-4 ; 
xi 26; 1 Cor. iii 17; viii 6; xv 28, 35; cf, Rom, vii 7 responden- 
dum est. 

simul (introducing a further consideration in a note): add to 


the examples given under adtendo, noto, etc. above, the following: 
Rom. i 16 simul et illos haereticos tangii; 1 Cor. xiiii 20 simul 
ostendit neminem . . .esse perfectum; 2 Cor. vi 14 simul ostendit 
neminem posse... iustuni esse...; 2 Cor. viii 15 ^\xn\x\ et ostenditur . . . 
omnia aequalia esse dehere; 2 Tim. iii 8 simulgwe ostendit Mosen 
ueritatis fuisse doctorem. 

Allusions to other passages in St Paul's Epistles are generally 
introduced by phrases containing the word alibi or the word 
superius : 

sicut alibi ait: Rom. viii 26; xii 10; xv 29; 1 Cor. vii 23; viiii 
27; 2 Cor. iiii 10; v 8; 1 Thess. ii 17. 

sicut alibi dicit: 1 Cor. xii 28; 2 Cor. iii 2, 18; viiii 7. 

et alibi ait Rom. xiii 13; sicut ipse alibi ait Rom. xi 28; et 
alibi Rom. viii 27; Eph. iii 18; sicut alibi 2 Cor. vi 2; 1 Thess. ii 3 
sicut ait alibi 1 Cor. vi 20; sicut alibi idem dicit Rom. xiii 11 
sicut aVibi. . .dicitur 1 Cor. iii 28; sicut ipse ait Rom. viii 39; Eph 
ii 6; sicut ipse alibi dicit 1 Cor. xi 23; unde et alibi ait Phil, ii 14 
ipse apostolus alibi dixit 2 Cor. iiii 4 ; unde et alibi dicit Gal. ii 1 1 
unde alibi ait 1 Cor. xvi 18; Col. ii 14; unde ait alibi 2 Cor. vi 4 
sicut et alibi simm nobis proponit exemplum dicens Rom. xv 2 
suum quoque alibi proponit exemplum dicens 1 Cor. viii 1 ; alibi 
appellauit 2 Cor. iii 7; 2 Thess. ii 3; unde et alibi... es^ appellata 
1 Tim. i 8; ut ibi 1 Cor. x 13; xv 50; sicut ibi Rom. v 12; 1 Cor. 
viiii 7; ut alibi dicit 2 Tim. i 7; sicut ibi dicit Rom. viii 27; ipse 
alibi dicit Rom. ii 13; viii 6; de quo alibi dicit Tit. ii 3; quo modo 
alibi dicat 1 Cor. xiiii 34; cf. 2 Cor. iii 6; quo modo scriptum est alibi 
Rom. xiii 7 (cf. Rom. i 30); dicente alibi apostolo 2 Cor. v 19; 
Paulum quoque ipsum alibi docere Gal. arg. ; iustitia, quam loricae 
alibi comjmrauit 1 Thess. v 8; quam idem apostolus alibi sine 
macida definiuit; dicit enim alibi Rom. v 7. 

Rom. ii 8 iam superius dictum est contentiosum... ; Rom. viii 8 
superius non carnem, sed opera accusauerit carnis; Rom. viiii 6 
quia superius dixerat dolere se quod...; Rom. viiii 29 superius 
scriptum est; 1 Cor. i 31 in superius conprehensis; 1 Cor. viiii 22 
in his quae superius memorauimus; 1 Cor. x 15 omnia quae su- 
perius conprehendi; 1 Cor. xii 30; xiiii 17; xv 10; 1 Tim. vi 8 
quod superius dixerat; 1 Cor. xiii 11 ilia quae superius me- 
morauit; 1 Cor. xvi 24 sicut superius ipse dem,onstrat; 2 Cor. 


V 12 quia superius dixerat; 2 Cor. vii 3 ut superius memoraiii; 
2 Cor. viiii 3 superius memoratos, qui. . . ; 2 Cor. viiii 9 sicut 
superius ait; 2 Cor. xi 16 iam enim superius dixerat; 2 Cor. xii 19 
sicut superius dixi; Eph. iii 1 huius (gratiae), quam superius 
memoraui; Eph. iiii 17 qiios superius obsecrauerat, hie...; 1 Thess. 
ii 8 conuersationem. . . , quam superius memorauimus; Tit. i 7 epi- 
scopuin, quern superius presbijterum nominauit: cf. supra Rom. ii 10; 
viii 2\ 

The introductions to scripture passages in general are, as is 
natural in the case of notes so brief, themselves also brief. For 
example, it is rarely, if ever, that an individual Gospel is named. 
Of course the author is not in this respect by any means unique, 
as the fourfold Gospel had long been an entity-. Again, the author 
speaks of propheta rather than the individual prophet, sometimes 
scriptura (sicut scriptum est)is all that is mentioned, and occasionally 
a scripture quotation is introduced without any warning at all. 
Illustrations of these facts can be obtained by consulting the index 
of scripture passages, and especially those that are cited diserte. 
The point need not be elaborated here. 

When a scripture passage is not quoted in full, it is customary 
to add et reliqua or et cetera to the quoted words: et reliqua'^ 
occurs: Rom. i 24; v 4; viii 23, 26; xii 12; xv 12; 1 Cor, i 11; 
xii 30; 2 Cor. vi 6; 2 Tim. i 12; Tit. ii 3: but et cetera is em- 
ployed: Rom. viii 9, 10, 36; xiii 1, 13 his; xv 3; 1 Cor. i 11; vi 1; 
vii 23, 31 ; viiii 27; x 6, 13; xiii 3; 2 Cor. iii 3; v 15; Eph. iiii 19; 

V 1; Phil, ii 14; Col. iii 13; iiii 6; 2 Tim. ii 20. 

(c) Characte7'istic Words and Phrases, alphabetically arranged. 

Words or phrases that are rarely found in the surviving literature, 
constitute by their presence an argument for common authorship, 
even if the examples adduced be few. If, however, the words are 
ordinary, they must occur a considerable number of times, or be 
used in a special way, must in fact be a sort of idiosyncrasy, to 
form any such argument. Examples of the latter sort bulk more 
largely here than those of the former, because, as has been already 

1 Cf . inferius Eom. ii 12 ; iii 12 ; 2 Tim. iii 9. 

2 Cf., for instance, the writer's Text and Canon of the New Testament (London, 
1913) p. 161. 

"■* On the adverbial reliqua see Archiv f. lat. Lex. ii (1885) 95. 


hinted, the author employs a very simple and correct style, and 
offers very little to the searcher after curiosities of expression. 
A further argument may be found in the use of ordinary words, 
if that use be conjoined with the absence or avoidance of equivalent 
words equally ordinary. 

a (ab) after the comparative : 1 Cor. xv 2 ; 2 Cor. iii 5 ; xi 5, 21 \ 

accessio (in the medical sense): 1 Tim. vi 4 diuersis accessio- 
nibus languet; Tit. ii 8 nullius adulationis accessione languentem. 

adiutormm (excessively rare in this author, though one of the 
commonest words in the late jteriod-): 2 Thess. i 11; 2 Tim. i 14. 

adsumo (in connexion with the Incarnation): Eom. viii 33 — 4 
secundum adsumpti hominis loquitur formam; 1 Cor. ii Sperfectam 
adsumpti hominis ?ia^t/ram... adsumptum hominem; 1 Cor. iii 23 
hie de adsumpti hominis /orma tractatur; 1 Cor. xi 3; Col. i 15 
secundum adsumpti hominis /ormam; 1 Cor. xy 4<H nostri generis 
adsumptus homo; Eph. i 21 ununi est iam cum, deo adsumptus 
homo; Phil, ii 9 adsumptus homo; Phil, ii 10 hominem... adsumptum. 

aemulor, etc.: Rom. xi 14 aemuler] Ut omni modo talem me 
exhibeam, ut me desiderent imitari; 2 Cor. vii 7 triplex est aemu- 
latio: aut imitationis aut inuidiae aut de qua agitur in praesenti; 
2 Cor. viiii 2 hie aemulatio pro imitatione ponitur; Gal. iiii 17 
aemulus et imitator potest et inimicus intellegi. Once or twice 
'aemulari' is defined as 'sectaril' 

aiiquanti (generally in contrast with omnes): Rom. viii 32: 
viiii 7 his, 26; xi 5, 14; 2 Cor. vi 4; xiii 13; Eph. i 3; vi 13; 

1 Thess. ii 2. 

aliquis (in negative or quasi-negative clauses) : Rom. vi 4, 22 
viii 36; xii 10; xiii 5; 1 Cor. ii 2; vii 10; viiii 12; x 20; xiiii 33 

2 Cor. i 4; vii 9, 12; viii 12, 21; xi 14, 21; xii 9; Gal. ii 20; vi 10 
Eph. V 15; vi 5, 18, 21; Phil, i 4, 13; ii 3; iii 1; 1 Thess. iii 3 
Col. iii 16; 1 Tim. ii 5, 8; 2 Tim. i 8; Tit. ii 5; iii 1; Philem. 14, 
23 (cf idiquando, 2 Cor, i 18; xi 6). 

alius = aZ^er^: Rom. xi22; ICor. ii8; vi 2; vii 5, 11; xv2, 18; 
2 Cor. xii 6. 

1 Also in the scripture of 2 Cor. xi 5, xii 11. 

'■^ Pel. has auxilium, suffragium. On adtiitorium see J. E. B. Mayor in Journ. 
Philol. XXII (1894) pp. 187 f. 

3 The Latin rendering of j'tjXwttjs in Tit. ii 14 is sectator. 

* See the index to J. E. B. Mayor's Latiii Heptateuch (Camb. 1889). 


alterutrum (as an adverbial phrase, = mm'cem^): prol. Rom.; 
Rom. i 19; xii 5; xv 14; 1 Cor. xii 25; xiiii 26, 33; Gal. v 26; 
Eph. iiii 16, 25-; 1 Thess. iiii 18. 

anathema: 1 Cor. xvi 22 anathema sit] qui earn non amant, 
anathematizabit, id est, ut illos abominetur et perdat; Gal. i 8 
anathema sit] Hoc est, abominabilis uobis sit. 

ante (adv.) is used more frequently than antea. Ante occurs 
Rom. i 30; iii 24; iiii 18; v 2; vi 13, 19; vii 8; viii 3; x 18 
xi 32, 34; xv 2, 4, 21; xvi 26, 27; 1 Cor. i 24; viiii 15; xv 24 
2 Cor. ii 3; xi 33; Gal. i 10; iii 28; iiii 9; Eph. ii 7; iii 21; Phil 
ii 9; 1 Thess. iii 4; 2 Thess. ii 2, 10; 1 Tim. i 14; iii 15 
2 Tim. i 10. Antea occurs: Rom. i 1; ii 29; viii 18; viiii 30; Gal. 
iiii 9; Eph. i 4; v 32; Phil, ii 5; Col. i 24. When the MSS vary 
between them, it is not easy to decide which is the true reading^ 

arefacio: Phil, iiii 10 iterum fiore honi operis florere coepistis, 
qui mei inmemores occupatione, non uoluntate, arefacti fueratis 
effecti; 1 Tim. vi 13 omnia quae uiuunt, etiam arefacta, per eius 
potentiam reuiuescunt. 

arguo (with a simple, usually personal, object): Rom. xi 7; 
xiiii 12; 1 Cor. viii 1; xi 31; xv 1 ; Gal. iiii 9, (20); 1 Tim. (iii 13); 
V 22; Tit. i 14, etc. 

aruspex: Rom. x 20 daemonia interrogahant per augures et 
astrologos atque aruspices idolorum; 1 Cor. xii 2 ducebamini] 
A magis uel ah aruspicibus idolorum. 

auctoritas (usually with reference to St Paul's apostolic 
authority): Rom. xiiii 4, 10; xvi 25; 1 Cor. i 1; v 4; 2 Cor. i 1 his; 
viii 10, 15; Gal. arg.; 1 Tim. arg.: i 1, etc. 

audenter: 1 Tim. v 22 utpossit audenter arguere delinquentes; 
Tit. i 6 ne non possit audenter corripere delinquentes; Tit. ii 7 ut 
audenter corripias delinquentes. 

baiulo: 2 Cor. iiii 7 thesaurum gratiae spir^talis in fragili 
corpore baiulamus; Gal. iiii 13 thesaurum gratiae in uasis fictilihus 

1 Cf. E. Lofstedt, Philologischer Kommentar ziir Peregrinatio Aetheriae (Uppsala 
and Leipzig, 1911) p. 337. 

- Also in the biblical text at this point (vg. inidcem) as in the TertuUian quota- 
tion (resurr. 45). 

Lofstedt, Fhilol. Komm. z. Peregr. Aetheriae, pp. 74 f. shows that the more 
popular language always preferred ante to antea. 


baptismum. There must often be doubt (in oblique cases) 
whether an author used baptismus or baptismum in the nominative 
singular. In Pelagius there is no certain proof of baptismus in the 
noni. sing., whereas there is clear evidence of baptismurn for that 
case. There are two examples only of baptisma, namely Rom. vi 3^; 
Eph. V 6. The forms of the o stem found are: — 

7iom. baptismum, Rom. v 15; vi 3. 

ace. baptismum, Rom. v 17; vi 22; 1 Cor. xv 49 bis; 2 Cor. 
viiii 2; Gal. iii 27; Eph. i 1; ii 5; 1 Tim. ii 15 (neut, as followed 
by quod); 1 Tim. ii 15 (again). 

abl. baptismo, Eph. ii 9; 1 Tim. vi 12. 

blasphemia (never blasphemium): Rom. i 18; v 2; 2 Cor. ii 7. 

breuiter: Gal. i 8 Breuiter omni uoluit praeiudicare personae\ 
Eph. V 3 omnia crimina hremter conp)-ehendit; Phil, iiii 8 ut omnia 
breuiter conprehendam; Col. ii 4 breuiter omnem sapientiam in ipso 
esse complexus sum; 1 Tim. v 10 Breuiter uniuersa conclusit; 
2 Tim. i lSformam...quam, a me breuiter accepisti. 

caducus: Rom. v 4 aeteniis caduca mutantur; 1 Cor. vii 38 
saecularibus et caducis. 

caerimonia : 1 Cor. vii 18 sine legis caerimoniis uiuat; 2 Cor. 
V 16 carnalite^^ circumcisum et carnales caerimonias obseruantem; 
Phil, iii 6 iustificationum et caerimoniarum purificationumque; 
Phil, iii 18 legis caerimoniis collocabant; 2 Thess. ii 4 
omnes legis caerimonias restaurare; Col. arg. Tie jt)er... legis caeri- 
monias seducantu?-^; Tit. ill caerimonias ludaeorum. 

calumnia, calmnnior, used exclusively, or almost exclusively, 
of the views or statements of heretics: the substantive, Rom. viii 
33 — 4; 1 Cor. viii 6, Tnuuere and commouere calumniam; also 
1 Cor. XV 27; 2 Cor. iiii 4; v 15; Phil, ii 5, sometimes in the 
phrase propter calumniam: the verb, Rom. ii 4; viiii 20; 1 Cor. 
xii 6; XV 24, 28. 

causa (in the sense 'subject,' 'matter,' 'case'^): Rom. iii 12 apostoli causam; Rom. iii 19 talia dixerat in 

1 Possibly the requirement of the clausula has to do with this example : if we 
substitute baptismi for baptismatis, we get a sentence ending with nine long 

2 This phrase legis caerimoniae occurs twice in Hieron. in Hierem. : see C.S.E.L. 
Lix 496. 

3 In the sense 'disease,' it occurs 1 Cor. xi 31. 


propria causa; Rom. iii 30 in isdem causis; Rom. iiii 17 ut prae- 
senti conueniat cans-Ae; Rom. v 1 pertractata causa; Rom. xi 20 nee 
adtendentes causam uel personas; Rom. xiii 1 haec c&usa.... prolata 
est; Rom. xiiii 5 in tali causa; Rom. xiiii 14 in talihus causis; 
1 Cor. i 10 Hinc iam causam contra dissensionem adgreditur; 
1 Cor. iiii 18 Hinc incipit causam fornicationis arguere; 1 Cor. vi 
18 grauare uult fornicationis causam; 1 Cor. vii 8 Incipit aliam 
causam de innuptis; 1 Cor. vii 15 ligatus in eiusmodi cansa.; 1 Cor. 
vii 38 Ita concludit uirginum causam; 1 Cor. xi 2 in hac causa; 
1 Cor. xi 4 Venit ad causam, quia et iiiri etc.; 1 Cor. xii 1 Causa 
incipit de spiritalibus donis; 2 Cor. viii 1 Causam inchoat de col- 
lectis; 2 Cor. viiii 2 usque ad finem causae; 1 Thess. iii 2 In hac 
causa dumtaxat; 1 Thess. iiii 11 quam causam in secunda ad eosdem 
plenius exsequitur; 2 Thess. iii 1 notandum in quibus causis... 
poscat; 1 Tim. ii 8 de hac causa... motam fuisse quaestionem ; 
1 Tim. V 23 ut...doctrinae causam... ci^re^; Tit, i 8 in causa 

cautela: 2 Cor. i 9 in nostra prudentia uel cautela; Eph. v 16 
uestra sapientia uel cautela. 

censeo : Rom. xiii 8 omnis homo proximus esse censendus est : 
1 Cor. iii 2 quid de illis censendum est, quibus...^'? 1 Cor. xiiii 3 
quicumque ergo haec habet, propheta esse censendus est; Eph. v 27 
maculati ab ea alieni esse censentur. 

cohortor with personal object and ad followed by a noun 
indicating a good moral quality: Rom. arg. eos ad pacem et ad 
concordiam cohortatur; Rom. xii 4 eos ad concordiam^ cohortatur; 

1 Cor. vii 29 habentes uxores ad continentiam cohortatur; Eph. 
v 29 (eos) ad continentiam cohortatur. (So hortor ad Rom. xv 8; 

2 Cor. i 6; Eph. vi 13; 1 Tim. iiii 2.) 

commenioro, with accusative of person and sometimes of thing 
also, or with object clause, in the sense, 'I remind': Rom. vii 8; 
XV 15; 1 Cor. xi 23, 26; xv 11; Eph. ii 11; 1 Thess. ii 1. (Also in 
ordinary sense and construction, Rom. i 7 ; iii 2; xvi 24; 2 Cor. ii 4.) 

commoneo Rom. i 7; viii 33 — 4; 1 Cor. iiii 17; vii 35; xi 26; 
2 Cor. viiii 2, 3; Phil, iiii 2; 2 Thess. iii 6, 12; 1 Tim. v 1; 2 Tim. 
arg.; i 6, 15; iiii 5; Tit. arg. 

^ Cf. 2 Cor. xi 17 quid de illis sentiendum est qui... ? 
2 Cf, Eom. XV 33. 


concludo: Rom. iii 18 in timore dei conclusit; Rom. v 1 qua 
ratione conclusa; Rom. vii 25 reccqntulat, ut concludat; Rom. viiii 19 
uestra propositio concludetur; Rom. xi 36 in ipso omnia conclu- 
duntur; 1 Cor. vii 38 concludit uirginum causam; 2 Cor. vi 2 
{dies) concluditur node iudicii: Gal. v 14 
possunt sermone concludi; Eph. vi 10 generali epistulam eochorta- 
tione concludit; 1 Tim. v 10 hreuiter universa conclusit. 

conparatio, in various phrases : adconparationem Rom. viiii 20 ; 

1 Cor. viii 12; 2 Cor. i 8; iiii 17; vi 10; xii 20; Eph. ii 21; Phil, 
iii 7; 2 Tim. ill bis: in conparatione (Augustine's favourite form) 

2 Cor. iii 10; in conparationem 2 Cor. vi 2 ; iuxta conparationem Rom. 
vii 4; per conparationem Rom. vii 2; xii 4; 1 Cor. viiii 7; xii 12. 

conprehendo ('I include') (see under breuiter) 1 Cor. i 31; 
x 15; XV 4; 2 Cor. i 1; Eph. iiii 19; v 3; Phil, iiii 1, 3; 1 Tim. 
iii 8, 16, etc. 

conprobo (=probo, 'I prove') Rom. iii 9; 1 Cor. x 22; xi 14; 
2 Cor. viii 23; Eph. i 15; 1 Thess. i 3; iii 12; 1 Tim. iii 5; v 20; 
2 Tim. i 6. 

consisto in c. abl. Rom. iiii 1; vii 23; viii 39; xv 13; 1 Cor. 
xi 22; Phil, iiii 18; 1 Thess. iii 8; 1 Tim. ii 2; Tit. ii 15. 

consuetudo, in various phrases: Rom. vii 3 secundum prioris 
6'1/ua'W consuetudinem ; Rom. xvi 23 secundum consuetudinem legis; 
1 Cor. i 22 ex consuetudine prophetarum; 1 Cor. iii 13 iuxta con- 
suetudinem scripturarum ; 1 Cor. v 5 habet consuetudinem scriptura; 
1 Cor. vi 18 illis per consuetudinem .leuissima uidebatur; 2 Cor. 
xi 28 ilia quae per consuetudinem leuiora esse uidentur; 1 Cor. 
vii 10 secundum consuetudinem ludaeorum; 1 Cor. viiii 6 aput 
ludaeos antiqua haec erat consuetudo, ut...; 1 Cor. viiii 13 aput 
ludaeos secundum consuetudinem ueteris testamenti; 1 Cor. xi 31 
putamus consuetudinis esse quod culpae est; Gal. iiii 4 per malam 
consuetudinem; Gal. v 17 caimalis consuetudo aduersus spiritale 
desiderium; Eph. ii 3 paternae traditionis consuetudo; 2 Thess. 
iii 13 boni operis consuetudinem; Tit. i 8 secundum consuetudinem 

contemno c. infin. Rom. ii 5; 2 Cor. iiii 4 (cf Thesaurus s.v. 
col. 637). 

contingo (3 sing, perf tense, generally of misfortune): Rom. 

contingo (3 sing, perf tense, i 
V 20; viiii 33; 1 Cor. i 16; vii 14. 

.s. p. 


contrarietas Col. i 20; 1 Tim. i 4. 

contrarius (in adverbial phrases): e contrario prol. epist. 
Rom.; Rom. viii 14; xiiii 8; 1 Cor. iii 12; vi 8; viii 8; xiiii 19; 
XV 33; 2 Cor. iiii 13; viiii 6, 13; xiii 11; Gal. iii 3: a contrario 
Rom. V 14. 

conuertor (depon.\ in various senses): Rom. ii 17 ; iii 29; iiii 5; 
viii 22; viiii 4, 12, 17; x 19; xi 8, 10, 11; xii 20; 1 Cor. vi 11; 
2 Cor. i 11; iii 6; Eph. arg.; Phil, iiii 22; 1 Thess. i 5, 8; 1 Tim. 
i 16; 2 Tim. iiii 13. 

corrigo (intr.^ of moral improvement): Rom. iii 26; 1 Cor. v 2 ; 
xiii 7; 2 Cor. ii 11, 13; vii 7; x 1; xiii 10; Gal. iii 4; Tit. iii 11; 
Philem. arg. 

credo (in the passive, with personal subject and infinitive, in 
the true classical manner: often credendus): Arg. omn. epist. 
ipsius niagis esse credenda est; Rom. i 1 quod... fecisse credendus 
est; Rom. i 8 simpliciter earn fidem laudasse credendus est; Rom. 
iii 28 per fidem dixisse credendus est; Rom. xii 15 nec.flesse cre- 
dendus est; 1 Cor. vii 16 semper amhigua in melius euenire eredendsi 
sunt; 1 Cor. xiii 3 magnis eam procul duhio rebus praetulisse 
credendus est; 1 Cor. xv 35 qui...CTed\i\xv totum reddere; Phil, 
iiii 21 proad duhio uerum dixisse credendus est; Philem. 2 Appia 
uel soror creditur eiusfuisse uel coniunx. 

denoto (with personal object): 1 Cor. xi 22; 1 Thess. iiii 12. 

deputo (always^ with accusative and dative): Rom. prol.; 2 Cor. 
i 23; Eph. i 2; 1 Thess. iiii 7; Col. ii 4; 1 Tim. v 6; vi 8. 

deseruio (generally with idolis and such like*): Rom. prol. 
idola, ..deseruistis; i 8 ut omnium gentium, diis-.-deserui- 
rent; Rom. viii 19 corrujytioin. . .deseruiie; Rom. viiii 21 Istrahel 
ihi ido\\B deseruierat; Col. iii 5 idolis... deseruire. 

diaconissa: Rom. xvi 1 etiam nunc in orientalibus locis dia- 
conissae mulieres in suo sexu ministrare uidentur in baptismo; 
1 Tim. iii 11 de his dicat, quas adkuc hodie in oriente diaconissas 
appellant; 1 Tim. v 9 fa7e5... diaconissas. 

1 Also reflexive Gal. iiii 9. See Linderbauer on Bened. reg. 2, 38. 
' Also active, e.g. 2 Cor. vii 12 (iu passive); Gal. vi 1; Phil, ii 12 (passive); 
1 Thess. V 14; 2 Thess. ii 4; 1 Tim. arg. 

3 Except Rom. prol. where in c. abl. for the dative. 
* Id good sense, Rom. i 9; 1 Cor. xii 25; xv 2. 


dialecticus : 1 Cor. i 22, 1 Tim. vi 21 artis dialecticae; 1 Cor. ii 4 
dialectici erant Corinthii; 2 Thess. ii 2 nee ser'mo dialecticae fallaciae 

digamus': 1 Cor. vii 39; 1 Tim. iii 12 2. 

dimico: Rom. xii 15; 2 Cor. vi 7; 2 Tim. ii 3; iiii 8. 

distraho ('I sell'): Rom. xv 25 omnibus suis distractis et ante 
apostolorum pedes depositis; 2 Cor. i 12 sapientia carnalis, quae 
mercede distrahitur; Gal. ii 10 omnia sua distrahentes ad apos- 
tolorum pedes pretia deponebant. (Possibly distraho was in 
Pelagius's copy of Acts at iiii 34; but Wordsworth and White 
give no authority for this synonym of uendo, for which see Mayor, 
Latin Hept. p. 67.) 

diuersitas: Rom. xi 5; 1 Cor. xv 39, 41, 42 his; 2 Cor. v 18; 
Gal. iii 28; iiii 25; Eph. i 23; iiii 30; 1 Thess. v 14 his; Col. iii 11 ; 
1 Tim. vi 4. In the first, eighth (?) and ninth examples, meritorum 
is the dependent genitive. 

diuersus: Rom. viiii 10, 17; xii 4; 1 Cor. xii 12; xv 2, 28 his, 
39, 42; Eph. i 10; ii 2; iii 6; iiii 7; Phil, ii 2; 1 Thess. iiii 11; 

1 Tim. iii 5, etc. 

doctor (a [Christian] teacher, sometimes probably the bishop'): 
Rom. xii 3; xv 27, 29; 1 Cor. i 2, 20; iii 10, 11, 12; iiii 6; vii 8; 
viiii 6; x 6; xi 19; 2 Cor. i 14; viiii 7; x 4; Eph. iiii 13; Phil, 
ii 2, 29; 2 Tim. ii 21; iii 3. 

dono: Rom. xii 6 donum non ex nostro, sed eo) donantis pendet 
arhitrio; 1 Cor. xii 11 cum hoc non in nostra, sed in donantis sit 
positum potestate. 

duplex, dupliciter: adj. Rom. vii 25; 1 Cor. vi 6; vii 26; 

2 Cor. i 17; viiii 12; Gal. v 16: adv. Rom. xi 28; 1 Cor. xi 29; 
Phil, iii 17; 1 Thess. v 13; Col. ii 13. 

efficio, used in the passive, as a mere synonym of fio : e.g. Rom. 
vi 0, 13, 19; vii 15; viii 17; xii 11; xv 8, 16; 1 Cor. i 2; xi 25, 
29; xii 13, 17; 2 Cor. iii 2; Gal. iii 27, 29; iiii 7; Eph. ii 4, 15; iiii 
10, 16; V 14; vi 5; Phil, i 14; iiii 10; Col. ii 13; Tit. ii 5. 

emendo (intransitive, of moral improvement, cf corrigo above): 
2 Cor. ii 3, 5; xiii 10; 2 Thess. iii 6, 14, 15; 1 Tim. i 20. (The 

^ See C. H. Turner's Monumenta luris Antiquissima t. ii pp. 16 f. 

^ Here also trigamus. 

3 Cf. Rom. xii 3 ; 1 Cor. x 6. 



active sense is frequently found, e.g. 1 Cor. vii 31 (in passive); 
2 Cor. ii 3 (in passive); vii 12 (in passive).) 

euidenter: e.g. 2 Cor. ii 11 ; Col. i 22; Philem. 6. 

exhibeo, in reflexive construction (me, se cf. the Scripture of 
2 Cor. vii 11, etc.): Rom. v 8; xi 14; xii 1; xv 24, 29; 1 Cor. vi 13; 
X 33; 2 Cor. viii 23; 1 Tim. iiii 12, etc. 

existo, in combination with causa: Arg. omn. epist. nouis 
causis existentibus; Rom. vii 13 causa... existit; 2 Cor. i 17 nulla 
maiore causa existente; Eph. v 15 incipiatis ei csaisa perditionis 
existere; Eph. v 22 nulla existente causa. (Similarly with occasio: 
Gal, V 15 dum alter alteri occasio perditionis exsistit.) 

exoro, not in the classical sense, 'I prevail on\' but as a mere 
synonym of oro: Rom. xv 30; Phil, i 4; iiii 7; 1 Thess. v 17; 
2 Thess. iii 1; Philem. S\ 

fiducialiter: 1 Thess. i 10; 1 Tim. iii 3; 2 Tim. ii 9; Tit. 
ii 13. 

finio: Rom. xv 33; xvi 1; 1 Cor. vii 29; xiii 8; 2 Cor. iiii 18. 

firmus, firmiter, firmo, used habitually of belief, faith: Rom. 
iiii 22 tarn perfecte et firmiter credidit; 1 Cor. iii 11 lesum, cui 
fide firmissima credidistis; 1 Cor. x 22 hinc fidem firmissimam 
conprobari; 1 Cor. xii 7 utilitatem] credentium, nt firmentur; 
2 Cor. iiii 14 firmiter tenentes hanc fidem; Gal. i 13 qiiam firmiter 
tenuerit ludaismum; Eph. arg. Ephesii crediderunt, quihus firmiter 
stantibus. . . ; Eph. i 15 quam firmiter teneatis fidem lesu ; Eph. iii 2 
firmiter retinetis me...accepisse...\ Eph. iii 17 ut in eius amore 
firmiter stetis; 1 Thess. iii 6 firmiter Christo credentes tenetis 
fidem; Col. i 23 firmiter /w^wra credatis. 

forma ('pattern,' 'example' in conduct): Rom. v 12, 16; viii 
33—4; 1 Cor. xv 22; 2 Cor. i 4; viii 8; Gal. iiii 3; v 14; (Phil, ii 5); 

1 Tim. V 23, etc. (A few other examples have been given earlier 
in the chapter ^) 

genero (metaphorically, like pario): Rom. xiiii 17; 1 Cor. x 13, 
23; 2 Cor. vi 6; Eph. iiii 18; Phil, ii 28; 1 Tim. v 11; vi 4; 

2 Tim. i 10. 

' Cf. R. Ogilvie, Horae Latinae (London, 1901) pp. 98, 226. On the opposite 
use, oro for exoro, see E. Lofstedt, PhUolog. Kommentar z. Peregr. Aetheriae 
(Uppsala & Leipzig, 1911) p. 41. 

2 Cf. Tert. resurr. 63 p. 124 1. 22 Kr. 

3 p. 69. 


gratias referred (particularly in passive construction): Rom. 
xiiii 6 bis; 2 Cor. i 11; viiii 8; Phil, iiii 6, 18; 1 Thess. v 18; 
Col. ii 7; iii 17 bis, 23. 

grauo (metaphorically), very often: e.g. 1 Cor. viiii 11; xi 11; 
xvi 2; 2 Cor. ii 10; Phil, i 17 (in Scripture 2 Cor. xii 13, etc.). 

habeo. Besides the ordinary use of habeo with the infinitive 
(Rom. ii 26 adferre, Rom. v 7 mo7-i, Phil, i 29 uinci), this author 
has instances of a vastly rarer use : Rom. iii 19 habent unde gloriari ; 
Rom. xiii 1 habeant quod timere; 1 Cor. xi 2 auctoritatem legis 
non habet quam proferre; 2 Cor. viii 14 non habet cui dare. 
(Contrast the classical, Rom. v 4 habemus ergo unde...gloriemur; 
Rom. X 13 habet unde. . .largiatur; Rom. xiii 3 bonus non habet quod 
timeat, etc.) Colloquial in origin, the use is found in certain 
Old- Latin texts of Scripture and in authors later than these I 
One or two examples from outside may be cited: Ps.-Aug. (Am- 
brosiaster) Quaest. uet. et nou. test. 112 §16 (p. 293 1. 9 ed. S.) non 
habet unde reus constitui; Hieron. Tractatus De Psalmis (Anecd. 
Mareds.iii(2)) 61 1 — ^non habebamus w6i requiescere^: Arnob.-Iun. 
Comm. in Ps. (Migne, P.L. Liii 526 a) habes unde uincere^. 
Further instances are quoted by Ldfstedt^. 

hie. Adverbial phrases with hie are rather characteristic: — 
ad hoc ('for this purpose') 1 Cor. i 21; 
„ „ followed by ut Rom. i 1; iii 26; iiii 6; 1 Cor. x 27; 
XV 41; 2 Cor. v 2; Eph. ii 17; 2 Thess. ii 14; 1 Tim. ii 2. 
ex hoc... quia Rom. v 5; 
„ „ ...quo Gal. iiii 6; 
„ „ ...ut Rom. XV 25; 
„ „ (simply) Phil, i 17. 
in hoc... quo 1 Cor. i 9; 2 Cor. vi 10; 
„ „ ...quod 2 Cor. vii 13; 
„ „ Phil, iii 19; 
„ „ ...ut 1 Cor. xi 3; (simply) Eph. i 14. 
per hoc ipsum...quod 1 Cor. x 12; 
„ „ (simply) 1 Cor. xi 26. 

1 Phrase also in Hier. epist. 22, 37 §1, Max.-Taur. 

2 See Ph. Thielmann in Archivf. lat. Lexikogr. ii (1885) pp. 63 f., in (1886) p. 532. 
^ A. S. Pease in Journal of Biblical Literature xxvi (1908) p. 118. 

* G. Morin in Eevue Benedictine xxvi (1909) p. 427; Etudes, Textes, Decou- 
vertes, t. i (Maredsous et Paris, 1913) p. 378. 

* Philolog. Konim. z. Peregr. Aetheriae p. 251. 


idcirco — quia: Rom. iiii 18; viiii 33; xvi 20; 1 Cor. i 2; 
2 Cor. iiii 13; 2 Tim. i 15, 18; iii 10. The reverse order: 1 Thess. 
iiii 14. 

idcirco — ut: Rom. vii 8; viiii 17; 1 Cor. viiii 15; xv 11; 
2 Cor. xi 33. 

idcirco — ne: Gal. iiii 9. 

ideo — quia: Rom. i 18; iii 4, 8; v 10, 12, 14; vi 19; x 19; 
xi 7; xiii 6, 8; xv 15, 24 bis; xvi 3; 1 Cor. viii 6; xi 21; xv 
24, 48; xvi 9; 2 Cor. i 12; ii 14; iii 5; iiii 13, 15; vi 2; xi 11; 
xiii 13; Gal. i 10; ii 12; iii 13 bis; v 11; Eph. iiii 18; Phil, iii 4; 
2 Thess. ii 16; 1 Tim. ii 15; 2 Tim. i 18; ii 3; Tit. i 1. [The 
reverse order of clauses is much rarer: Rom. vii 9; xvi 13; 1 Cor. 
viii 8; Col. iii 12]: ideo — quo Rom. xiiii 14. 

ideo— ut: Rom. iiii 24; xii 17, 20; xvi 20; 1 Cor. x 16, 30; 
xi 26; xii 21; 2 Cor. i 2, 7, 24; ii 4, 5; Gal. iii 13; iiii 14; Phil, 
iii 10; Col. ii 4; 1 Tim. vi 17; 2 Tim. ii 10; iii 17; ideo— ne Rom. 
iii 13; 1 Cor. vi 10; x 11; 2 Cor. i 1; x 10; Gal. iii 20; 1 Tim. 
V 14. 

imago contrasted with ueritas: Rom. ii 26 uisibilia imago 
sint inuisibiliiim, et inuisibilia ueritas sivt uisibilium. ..imagine 
non indiget ueritas, imago auteyn indiget ueritate; 1 Cor. vii 31 si 
talis est imago, ipsa ueritas qiialis est! Eph. ii 21 ut multo maiorem 
munditiam, et sanctitatem habeat ueritas quani imago; Col. ii 16 
imagine opus non est ueritate praesente. [Figura is similarly 
contrasted with ueritas; 1 Cor. v 8; vii 31; x 6.] 

impedio with the dative, a rare construction, is found Rom. 
vii 2, and possibly 1 Thess. ii 16. (The ordinary accusative occurs 
at Rom. viii 31 ^) 

impugno: Rom. v 15; xv 25; Eph. vi 13; Phil. arg. (cf im- 
pugnatoi' Rom. vii 12). 

incautus: the ace. pi. incautos thrice ends a sentence; 2 Cor. 
ii 11 ut sub specie iustitiae fallat incautos; 2 Cor. xi 14 ne forte... 
decipiat ac fallat incautos; Eph. iiii 14 « recto Jidei cursu detorquet 

incorruptibilitas: Rom. i 4; Col. i 18. 

indebitus, indebite : Rom. iii 24 morti se ille indebite tradidit ; 
Rom. V 6 indebite pro nobis mortuus est; Rom. v 8 indebite ali- 
quid praestatur. . .quid tarn indebitum qiiam ut sinepeccato dominus 
^ See also Eom. xv 32; 1 Cor. viiii 25. 


pro seruis impiis viorereturl Rom. xii 20 tua'ni indebitam miser i- 
cordiaTn; Gal. iii 13 indebito maledicto eius nostrum dehitum 
conpensatum est. 

indigeo is his favourite verb^ to express the idea, 'I need/ and 
is used in various constructions: (c. gen.) 1 Cor. iiii 8; 2 Cor. vi 10: 
(c. abl.) Rom. ii 26 bis; xi 33, 34; xii 13; 1 Cor. xii 25; Gal. iii 25; 
vi 1: (c. infin.) Rom. xv 24; 1 Cor. v 10; Gal. i 18; 1 Thes.s. 
iiii 9; 2 Tim. i 6: (absol.) Rom. xii 8; 2 Cor. xii 10; Gal. vi 10; 
Eph. iiii 28; 1 Thess. iii 12: (c. ace.) Rom. i 11. 

inebrio (metaphorically, perhaps under the influence of Hierem. 
xxxi (xxxviii) 24 (25)): Rom. vii 15 quasi inebriatus consuetudine 
peccatorum ignoro quid facio; 2 Cor. ii 17 adulationis mollia uerba, 
qtdbus homines non tarn inebriati /er wean ^ quam delectati tepescant; 
Eph. iiii 26 ira non inebriet mentem; 1 Thess. v 6 et curae ine- 
briant mentem; Col. iii 8 ira est quae inebriat mentem. (In the 
literal sense, 1 Cor. xi 19.) 

ingratus c. dat.^: Prol. epist. Rom. semper his omnibus 
(beneliciis) extitistis ingrati; 1 Cor. xi 26 ut beneficiis eius non 
exsistamus ingrati; Gal. i 4 ostendit beneficia Ghristi, quibus 
existebant ingrati; Gal. ii 21 non debeo esse illi ingratus, qui me 
tantum dilexit, ut, etc.; Eph. arg. ut tantis beneficiis non sint in- 
grati; Eph. i 2 si ei (gratiae) non sitis ingrati; Eph. ii 11 ut non 
sint ingrati beneficiis largitoris; Tit. ii 9 ne homini existatis 
ingrati. So in Fulgentius, etc. cf Friebel, Fulgentius [Paderborn, 
1911] pp. 19 f 

inpossibilitas (in the active sense, 'impotence'): Rom. iiii 20; 
Gal. iiii 24. 

inremediabiliter: Rom. xi 11 non penitus et inremediabiliter 
ceciderunt; 2 Tim. ii 17 ne per aures inremediabiliter uulnerent 

inrogo: Rom. viiii 2 quidquid ei inlatum' iniuriae fuerit, 
Christo similiter inrogari; 1 Cor. vi 8 cum inlatam (iniuriam)... 

1 Egeo also occurs, e.g. Rom. xii 4; 1 Cor. xii 24; 2 Cor. vi 10 bis; viii 8, etc. 

2 Pelagius appears rather to affect adjectives followed by the dative, e.g. Rom. 
xi 8 incredulus uerbis; 1 Cor. xv 36 injidelis promissionibus; 2 Cur. iv 1 idoneus 
officio. Tert. (after Verg. ) uses ingratus with the genit., cf. Mayor on Tert. Apol. 
40 p. 118 1. 22, LSfstedt, Krit. Bemerkungen zu Tert. Apol. (Lund, 1918) pp. 92 ff. 

^ The classical phrase infer re iniuriam occurs in Rom. xii 10 and elsewhere, as 
well as in the above three passages. 


patienter sustinere deberetis, uos e contrario non solum non 
suffertis, sed etiam non facientihus inrogatis; Gal. v 22 inlatas 
iniurias sustinere patienter... ?i?i./^i iniuriam inrogare. 

insensibilis: Rom. i 16 insensibile aurum (as used for images 
of gods); 1 Cor. x 20 idolum insensibile; Eph. ii 12 idola insen- 

is in various adverbial expressions^: eo quod Arg. omn. epist. 
Rom. (i 3); ii 26; iii 11; iiii 13; viii 31; viiii 2, 20; 1 Cor. ii 1; 
vi 5; X 27; 2 Cor. i 12; vii 9; xi 30; Eph. i 3; Phil, ii 26: pro eo 
quod Rom. i 21; 2 Cor. i 11; 1 Tim. ii 1: ex eo quod Rom. i 32; 
iiii 2; Eph. i 1: in eo quo Rom. viiii 10; 1 Cor. x 1; in eo quod 
Phil, i 9: ad id quod 1 Cor. xii 18; 2 Thess. i 11; (ad quod 
1 Cor. xii 14): ab eo quod 1 Cor. xv 241 

iubeo c. dat. Rom. v 14; 1 Cor. i 13. 

laesio : 1 Cor. xvi 1 1 animi laesionem ; 2 Cor. vi 4 omnis laesio 
tribulatio est. 

legalis: Rom. viii 33 — 4 mandata; 1 Cor. vii 18 opera; 1 Cor. 
viiii 21 doctrinse. 

libertas arbitrii, the two words being separated; Rom. xi 8 
ne libertas scilicet tollatur arbitrii; 1 Cor. iii 23 ad libertatem 
referre arbitrii. Cf. 1 Cor. xv 10 ut liberum seruaret arbitrium. 

ligo, participle ligatus, metaphorically, of persons: 1 Cor. vii 
15, 27 {his); Eph. iii 1; (iiii 5). 

littera, in the expression lex litterae, meaning the written 
Law as contained in the Old Testament^: Rom. iii 20, 21; vii 6, 8; 
1 Cor. viiii 21; Gal. ii 19. 

locus, in abl., contrasted with some other relation*: 2 Cor. vi 
17 exite. . .separamini]. Actu uel conuersatione uel familiaritate, 
non loco; Eph. v 31 relinquet homo patrem et matrem 
suam]. Amore, et si non loco. 

1 It may be noted that Pelagius uses hie, ille and is alike, as antecedents to qui. 

2 In scripture passages we find eo quod (2 Cor. v 4), e.r eo quod (2 Cor. viii 11). 
On this type of phrase see Yearns Work in Classical Studies for 1916 pp. 44 f. 

^ The only instance of this phrase known to me, outside Pelag., was Hieron. 
in Hierem. vi26 §4 (C.S.E.L. lix p. 404 11. 25 f.) legem litterae lege spiritus commn- 
tatani: Dr Alfred J. Smith observes it also in Orig.-Ruf. in Eom., as his list of 
examples in Journ. Theol. Stud, vol.xx (1918 — 19) pp. 133 f. shows; and J. P. Naish 
in Gaudentius. 

* Cf. tempore (below), similarly used. 


macule (metaph.): Rom. xv 17; 1 Cor. iiii 4; viiii 12; Eph. 
V 27; 1 Tim. vi 14. 

mensura: Rom. i 30 qui effertui- super mensuram suam; 2 Cor. 
X 12 (cf. scripture text of v. 13) noti excedimus mensurae nostrae 
terminos; 1 Tim. iii 3 qui omnia facial cum mensura. 

merces (much commoner than praemium^, and in most passages 
used of the future, heavenly reward): Rom. ii 7; iiii 4; viii 28 
xii 1; xiii 4; xiiii 5; 1 Cor. iii 8, 19; vii 7; xv 31; 2 Cor. i 12 
viii 17; viiii 12; x 15; Gal. v 5; vi 9; Eph. vi 7, 8; Phil, i 20, 28 
ii 14; iiii 14; 1 Thess. ii 2; 1 Tim. iiii 8; 2 Tim. ii 6; Tit. ii 9 
Philem. 14. 

mereor c. infin.^Rom. viii 14, 17 bis; x 19; xii 1; 1 Cor. vi 14; 
vii 5; viiii 23; xv 22; 2 Cor. v 2, 4; Gal. iiii 20; 1 Thess. iiii 3 
(but mei^eo, 1 Cor. ii 10; Gal. i 10; 1 Tim. vi 2, etc.). 

monstruosus : 2 Cor. xii 21 inpudicitia... quasi monstruosa 
turpitudo; 1 Tim. iiii 5 monstruosa nescio qua, praedicatio. 

moralis: 1 Cor. xiiii 6 uos moralibus doceam institutis; Gal. 
iiii 24 moralia praecepta... moralihns monitis; Gal. v 5 moralis 
iustitiae: Gal. v 7 moralibus disciplinis; Gal. v 14 legis ipsa 
moralia; Eph. arg. moralia... instituta; Eph. ii 15 moralia sola 
decerne7is; Eph. iii 21 vaoraMa. . .trader e instituta. 

munio beginning the sentence: 2 Cor. x 4 Muniunt et circwn- 
dant doctrinam suam falsi doctores astutia argumentisque, quae 
aries ajjostolicus destruit...; Eph. vi 14 Munit non solum pectoris 

nee non et is much rarer in this author than in some other 
late prose authors. I have noted only the following examples: 
1 Cor. viii 1; Eph. vi 14; 1 Tim. vi 17. 

necesse est seems always to take ut with the subjunctive, e.g. 
Gal. V 3. 

nitor c. infin. Rom. v 15 ; 1 Cor. vii 34; xiii 3; Gal. i 7; iiii 29; 
1 Thess. iiii 6; Col. iii 10; 1 Tim. i 10; (c. abl. 1 Cor. viii 1). 

nobilitas in certain phrases: 1 Cor. iiii 10 uos nobis etiam 
nobilitatem terreni^ generis uindicatis; 2 Cor. xi 18 in nobilitate 

1 Which occurs, e.g. 1 Cor. vii 26, 29, 40; 2 Cor. vi 10; Phil, ii 14, 18; iii 15; 
1 Tim. iii 13. 

2 See Study of Ambrosiaster p. 118, and add Tert. Apol. 33, Orat. 16, etc. Aug. 
^ Cf . under terrenus below. 


terreni generis gloriari carnale est; 2 Cor. xi 21 si filii dei de 
terrena nobilitate se iactent; Eph. iiii 5 nolite uohis de terrena 
nobilitate aliquid adrogare, ne ueram gloriam amittatis. 

nomino greatly predominates over nuncupo^, uoco^: Rom. i 8, 
9, 13; ii 10; vii 8; viii 27, 39; xvi 5, 23; 1 Cor. i 2; iii 10; x 16; 
xilO; xii6, 28; xv50, 51; xvi 19;; vi 15; xi3; xiiil3; 
Gal. i 10; Eph. ii 14; iii 1; v 3; 2 Thess. ii 16 his; Col. i 23; iii 5 
bis, 12; 1 Tim. ii 5; Tit. i 7,8. 

noui = scio^: Rom. i 18, 27; ii 1; iiii 19; v 5; viii 25; xv 15, 
30; 1 Cor. i 11; vi 5; vii 7, 38; viii 8; x 27; xiii 2; 2 Cor. i 13; 
iii 4; v 11; viiii 5; xi 11; Gal. ii 14; Eph. i 17; iii 18; Phil, i 25; 
ii 22, 28; iiii 5, 8; 1 Tim. iii 7; 2 Tim. i 8; ii 18; iii 9; iiii 8. 

nuUus in the phrase in nuUo = nulla in re^: Rom. vi 20; 1 Cor. 
iiii 8; xi 27; 2 Cor. ii 9; Gal. ii 8; 1 Tim. iii 11. 

obiectio: Rom. viiii 14 quibus ewemplis per breues obiectiones 
respondens ostendit ita intellegi non debere; Rom. ^ 14 de gentibus 
obiectio ludaeorum; Eph. v 31 obiectio carnaliter amare uolentium ; 
1 Tim. ii 4 illud hoc loco soluitur de induratione Pharaonis et cetera 
huiusce modi obiectio quaestionum. 

obliuio, in the phrase in obliuionem ire: Rom. iii 20 in 
obliuionem ierat lex naturae; Rom. vii 8 paene lex in obliuionem 
ierat naturalis; 1 Cor. x 6 ut... in obliuionem nobis eant Aegypti 

occasio, generally of the opportunity to do evil': Rom. vi 18 
non solum peccata, sed etiam occasiones auferri docuit delictorum ; 
1 Cor. vi 12 cum peccata non sint, occasiones possunt capere delic- 
torum; 1 Cor. viiii 12 aduersariis occasione accepta deuorantibus; 

1 Cor. X 23 per se non habent peccatum, sed non semper expediunt, 
quia occasionem non nuniquam generant delinquendi; 1 Cor. x 30 
^t^...occasionem demus infidelibus blasphemandi; 2 Cor. xi 12 ideo 
tollit occasionem apostolus, ut...; Gal. v 15 alter alteH occasio 

1 Nowhere found, I believe; hut appello occms : Eom. il4, 29; iilO; iii 12, 19; 
vii 8; viii 2, 5, 6; xi 6; xiii 6, 7; xvi 3; 1 Cor. iii 3; viiii 21; xv 26, 48; 2 Cor. 
iii 7; vi8; Gal. iiii 3, 29 ; Eph. arg. ; vi 12; Col. iii 5; iiiil; ITim. i4, 8; iii 11; 

2 Tim. ii 8. 

2 Eom. viii 3 nomine uoco. 

3 Scio, e.g. 1 Cor. xii 28. On scio — noui, see Mayor on Tert. Apol. c. 5 p. 175 
11. 24 ff. 

■* Mayor in Journ. Philol. xxii (1894) p. 195. ^ To do good, 2 Cor. ii 13. 


perditionis exsistit\ 2 Thess. iii 9 necui occasionem auaritiae uel 
otii...'praeheremu8\ 1 Tim. i 9 qui didicerunt etiam occasiones 
fugere delictorum; 1 Tim. ii 9 non debent occasionem dare con- 
cupiscentiae; 1 Tim. v 4 ne eis ipsa occasionem det saeculo seruiendi. 

opto (in present tense ^): Rom. i 7 ut ea optet in nobis integra 
permanere; Rom. vii 19 etiam cum non optat, incurrit; Rom. xii 18 
oiptaintes...conuersionem eoruni...; Rom. xv 25 quibus oblationem 
suam esse optat acceptam; Gal. vi 18 optat ut gratia cum eis 
domini, non legis opera comitentur; Eph. i 17 optat ut agnoscant 
deum; 1 Thess. iii 10 quis, cum famam compererit bonorum, eos 
uidere non optet? 1 Tim. v 15 ne eum putaremus optare; 2 Tim. i 4 
tristitiani... , quam tua opto praesentia releuari. 

paganus: Rom. i 16; 1 Cor. xv 2; Gal. vi 10; Eph. vi 9; 
Phil, i 1 ; Col. iiii 6. 

parco, used somewhat baldly, with the dative: Rom. ii 2; 
1 Cor. X 1, 11 bis; Eph. v 6 bis; 1 Thess. ii 15; 1 Tim. v 20. 

pasco : Rom. xiiii 4 tres pueros leguniinibus pastos ; 1 Cor. x 27 
non diuites pascendos esse, sed debiles; 1 Tim. iiii 8 ipsa uidua... 
pasta est. 

passibilis: 1 Cor. ii 8 bis; Eph. iiii 30. 

perficio, with personal object, e.g. Eph. iiii 12; very frequently 
in the participle perfectus: Rom. v 4, 5; vi 2, 14; 1 Cor. i 10; 
ii 8; vii 7; xiii 2, 3, 9; xv 45; 2 Cor. i 4; ii 8, 9; iii 7, 8; v 8. 9, 
16; viil; viiil4, 15; Gal. iii 11, 24, 25; v5; Eph. i 23; Phil.ii8; 
iiii 19; 1 Thess. arg.; i 7; iiii 4, 9, etc.: occasionally (illogically) 
the comparative perfectior is found : Arg. omn. epist.; Rom. xiii 11 ; 
1 Cor. vii 38; xi 17; Phil, iiii 1. The adverb perfecte also occurs, 
Rom. iiii 22, 24; 1 Cor. xi 24; xiii 12; 2 Cor. ii 3; viiii 13; Eph. 
i 17; Phil, iii 10; 1 Thess. iii 2; 2 Tim. ii 4; Tit. ii 12, and the 
substantive perfectio, 2 Cor. viiii 12; Eph. iiii 12, etc. 

persona (in non-theological sense): Rom. iii 2 ex cuius -persona 
respondetur; Rom. vii 7 in persona eius hominis loquitur, qui...; 
Rom. vii 25 unde pr-obatur quia ex alterius persona loquatur 
apostolus, non in sua; Rom. viii 30 discretio non in personis, sed 
in tempore est; Rom. viiii 20 quibusdam uidetur et hoc adhuc ex 
ipsorum persona dicere, quia...; Rom. viiii 26 eos, qui haec non 
ex apostoli, sed ex ludaeorum persona did puta7it; Rom. viiii 30 

1 The future occurs Eph. vi 5; perf. subj. Phil, iii 18. 


si superiora ex persona apostoli dicuntur; Rom. xi 20 nee adten- 
dentes causam uel personas; 1 Cor. xv 35 ipse sibi ex contradi- 
centium persona proponit; 1 Cor. xv 55 propheta ex persona 
iustorum loquitur; 2 Cor. ii 10 ego dono, non in mea persona, sed 
Christi, qui dixit...; 2 Cor. iiii 6 in persona Christi nos hoviines 
scientia luminamus; 2 Cor. xii 2 de se humilitatis causa in alterius 
persona loquatur; Gal. i 8 breuiter omni uoluit praeiudicare per- 
sonae; Gal. ii 6 nee persona (praeiudicat) labori; Eph. iiii 19 in 
libro Sapientiae dicitur ex persona eorum, qui. . . ; 2 Tim. iiii 8 non 
personis meritum, sed labori debetur. 

portendo: Rom. i 4; 1 Cor. xi 25; 2 Cor. iii 3. 

postmodum (never postmodo) : 1 Cor. x 4 ; 2 Cor. iii 3. 

praeiudico: e. dat. Rom. viiii 6, 12; Gal. i 8; ii 6; 1 Tim. i 15; 
V 21: absol. Col. iii 11. 

praeposterus : 2 Cor. viii 3: praepostero (verb) Rom. ii 24, 

praesens. The adverbial phrase in praesenti occurs frequentl}^ 
in praesentia once^ only, 2 Cor. x 7: neither in praesens, nor 
Jerome's favourite, in praesentiarum, occurs. The examples of 
in praesenti are: — Rom. i 32; ii 4 bis\ v 13; vi 22 bis; vii 9 
viii 6, 39; xi 34; xii 6; 1 Cor. xiii 12, 13; xv 2 bis, 19; 2 Cor. i 4 
vi 4; vii 7; xiii 4; Gal. vi 9; Phil, iiii 1; 1 Thess. ii 2; iii 12 
Col. i 10; ii 17; iii 4; 1 Tim. iiii 8, 10; Tit. i 5. ' 

praeualeo: absol. Rom. v 15; 2 Cor. xi 16; xi 33; Gal. iii 9; 
c. infin. Eph. iii 8; Phil, i 10; Col. iiii 2; c. dat. Rom. xvi 20; 
2 Cor. xi 30. 

principor: 1 Cor. xv 24 nobis se sequentibus principantur ; 
Eph. iii 10 qui rebus caelestibus. . .-principantur. 

prior: Rom. i 4 ut prior omnibus. . .resurgeret; 1 Cor. xv 3 uel 
a lege uel a prioribus; Eph. i 12 nos apostoli uel ludaei, qui priores 
credidinius in Christo ; Eph. iii 5 sciebant quidem prophetae priores 
gentes esse uocandas; 2 Thess. i 11 priores inuitati non erant digni; 
1 Tim. ii 14 posteriores infactura sunt et priores in culpa; 2 Tim. 
i 5 Quia prior credidit. 

profectus (subst.)- Rom. i 8; xv 14; xvi 13; 1 Cor. i 4; ii 5 ; 
iiii 14; x 27, 29, 33 bis; xiii 11; xiiii 6, 19; xvi 14, 17; 2 Cor. vi 

1 Gal. iiii 18 is, of course, diffei-ent. 

2 See the passages indicated earlier in this chapter, p. 70 : Dr Alfred J. Smith 
shows that both proficio and jjrofectits occur frequently in Origen-Rufinus in Rom. 
(seeJourn. Theol. Stud. vol. xx (1918—19) p. 148 n.). 


11 bis; vii 4, 9, 13, 14; viii 24; Gal. ii 11; iiii 21; Phil, ii 17; 
iii 13; 1 Thess. i 5; ii 20; iiii 1; 2 Thess. iii 1; 1 Tim. iiii 15 1; 
Tit. i 10. 

proficio is, as in Ambrosiaster-, very frequent: (a) absol. Rom, 
XV 29; 1 Cor. i 19; v 10; xi 17; xiiii 5; xv 31; 2 Cor. ii 5; iii 7 
vi 1; vii 7; Gal. iiii 11, 17; v 12; vi 4; Eph. i 10; iiii 2, 16 
vi 21; Phil, arg.; iii 13; 1 Thess. arg.; i 7; ii 1, 8, 12; iii 9 
Col. iii 23; 1 Tim. iiii 12; v 1 ; (h) c. ad (in) et ace. (in bad sense) 
Rom. iii 26; 1 Cor. xi 17; xvi 11; Eph. iiii 28; Phil, i 10; 1 Tim 
vi 1; (c) c. ad (in) et ace. (in good sense): Rom. vii 4, 10; xi 15 
xiii 4; xv 4, (29); 1 Cor. xiii 9; 2 Cor. iii 18; iiii 17; Gal. i 7 
iii 10; Eph. iiii 28; vi 5; Phil, i 10; 1 Tim. iiii 8, 15; {d) c. dat. 
Rom. xiii 4; 1 Cor. xiiii 10, 19; Col. iiii 3; 1 Tim. ii 2; (e) c. con- 
tra: Eph. iii 9; (/) c. de: Tit. iii 9. 

propono: Rom. vii 15; viiii 14, 17, 30; xv 2; 1 Cor. viii 1; 
X 1; xiiii 6; xv 2,S5 bis; 2 Cor. arg.; vl6; vi 14; viii 19; 2 Tim. i 9; 
Tit. i 2; iii 3: the usual object is exemplum. 

proprie: Rom. ii 22; 1 Cor. i 2; iii 3, 18; viiii 10; xii 6; 
2 Cor. ii 15; vil6; Eph. vi2; Phil, iii 9, 11 (?); lTim.iil5; iiii 6; 
vi 16, etc. 

prouoco^: Rom. i8; xii 15; xiiii 15; xvl4, 27, 29; 1 Cor. i 4 
vi 6; viiii 2, 21; x 33; xi 2; 2 Cor. vi 11 bis; vii 4; viii 1, 22 
xiii 4; Gal. v 26; Phil, iii 13; iiii 18; Col. iiii 5; 1 Thess. arg. 
1 Tim. iii 1, 2; 2 Tim. i 16; iiii 6; Tit. iii 2. 

pulchre: Rom. i 29 pulchre homicidium inuidiae sociauit; 
Rom. iii 29 pulchre moduni seruauit in uerbis; Rom. xv 33 pulchre 
in pace finiuit; 1 Cor. x 4 pulchre dixit: 'consequenti petra'; 
1 Thess. i 9 pulchre ad deum iierum et uiuum afalsis dis et mortuis 
conuersi esse dicuntur. 

puto (in passive forms): e.g. Rom. ii 4 putatur res humanas 
minime curare; Rom. v 13 ita putabatur esse peccatum; Rom. 
viiii 10 Rebecca putatur prima geminos edidisse; 1 Cor. i 25 
quod stultum putatur dei; 1 Cor. xv 6 ne...putarentur esse men- 
titi; 2 Cor. vi 3 ne .. .nostrum uitium putetur esse; 2 Cor. xii 20 

1 In the scripture text of this verse the word occurs. 

2 Study of Ambrosiaster pp. 129 — 132. The 'good' sense with ad (in) is fouud 
also in Tert. Ambr. Aug. Vincent. -Lirin. 

3 See earlier in the chapter, p. 70. 


ista nee putantur esse peccata; 1 Cor. xv 31 non statim per puerum 
iurasse recte pu tabor. 

quale est, exclamatory: 1 Cor. vii 33 quale est uxori magis 
uelle placere quam domino! Eph. v 22 quale est ut nouae uitae 
praedicator, nulla exsistente causa, hoc doceret quod naturaliter 
possidebant^ ! 

qualitas: Rom. iiii 25 in ea qualitate, qua mortuus est, neces- 
sario resurgens apparuit; Rom. xi 24 radix {solet) ramoruvi (uim) 
in suam uertere qualitatem; 2 Cor. xi 20 ludaica superhia-.-de 
generis qualitate; Gal. iiii 3 {elementa) fuerant eorum infirmitati 
necessaria pro temporis qualitate; Gal. iiii 25 de qualitatibus 
locorum unit intellegi diuersitatem testamentorum; Eph. iiii 29 
qualitatem loquendi monstrauit; Phil, iii 7 utrumque (i.e. aurum, 
aeramentum) pro temporis qualitate necessanum; 1 Thess. i 9 ei 
nostrae constantiae et uestrae conuersionis omnibus nota est quali- 
tas; 1 Tim. i 8 {lex bona) ut a deo promulgata pro temporis 
qualitate; 1 Tim. iii 13 qualitatem ministerii praemium, promereri 

quanto magis exclamatory, occurs with a frequency usual in 
this period: e.g. Rom. i 20; ii 2; v 4, 9, 10; xi 4, 12; xv 7; xvi 
19; 1 Cor. i 12; iii 5; vi 20; vii 31; viiii 12; x 11; xi 17,27,28 
xiii2; xiiii7; 2 Cor. iii 3, 11; viii 10; viiii 4; xi5; xiii3; 1 Thess 
iiii 16; Col. iii 8, 13; 1 Tim. iii 2; v 19; vi 2; 2 Tim. iiii 15 
{quanto minus Rom. x 16; xi 21 etc.) 

quantuslibet: 2 Cor. iiii 17; quantumlibet (adv.) Col. ii 19. 

quantumuis c. subj.: Gal. iiii 30; Eph. v 6. 

quasi si: Rom. v 20; viii 3; 2 Cor. vii 9; 2 Thess. i 6. 

qmso^ue = quisquis, as fairly frequently in late authors: Rom. 
xiii 10; 1 Cor. i 1; v 5; xi 25; 1 Thess. iiii 6; 2 Tim. iii 7. 

non quo, introducing a rejected reason or view in the sub- 
junctive, as in the best classical Latin ^: Rom. xiiii 14; 1 Cor. vii 
13; xii 23; 2 Cor. i 24; Gal. v 16; 2 Tim. iiii 17; where a second 
clause is added in contrast, it is generally in the form of a principal 
clause introduced by sed, but there is one example of the full- 
blown classical phrase non quo... sed quia, 1 Cor. i 21. 

1 Also relative, Rom. v 4 ; viiii 17 tale est... , quale si{quis)... . For the exclama- 
tory use, see Tert. de fug. in persec. 5, with Oehler's note, and Hoppe, Syntax u. 
Stil des Tertullian (Leipz. 1903) pp. 68, 82. 

2 Jerome and Augustine also preserve this use. 


quoad usque: 1 Cor. xv 32; Col. i 24. 

rationabilis : Rom. xvi 20; 1 Cor. viiii 21; Eph. i 11; Phil, 
ii 15; Col. i 23; rationabiliter: Rom. xii 1; 1 Cor. xii 8; Eph. 
vi 20; Col. iii 16; 1 Tim. iiii 6. 

reddo uicem^: Rom. xii 17, 21; 1 Cor. iii 18; iiii 10 bis; Eph. 
iiii 2, 31; 1 Thess. v 15 bis; Col. iii 15; Tit. iii 2; (rependere 
uicem 2 Cor. v 14 ^j. 

replico: Rom. prol. sed quid antiqua replicamus... ? 1 Cor. 
viiii 1 suam illis replicat ybi'mam, quod etiam licita contempserit; 
Eph. ii 1 incipit collata beneficia replicare, 

ritu: Rom. v 12 qui humano, nan caelesti ritu uiuebant; xiiii 14 
qui adhuc ritu ludaico aliquid. . .arbitratur inmundum; 1 Cor. ii 14 
quia animalium ritu uersatur; 1 Cor. vii 39 tantum ut infideli uel 
infidelium ritu non nubat; 1 Cor. xv 47 non naturae fragilis ritu, 
sed diuinae maiestatis nutu et conceptus est et eniocus; Eph. iiii 21 
ut...gentili ritu uiuatis. 

saepe is entirely absent, I think, as from many other late 
authors, being replaced hj frequenter, etc. 

sane occurs with frequency, sometimes as the first word of its 
clause, far oftener as the second: (a) in the first place: Rom. iii 30; 

1 Cor. xii 8; 2 Cor. vii 7; Gal. vi 1; Eph. iii 1; {b) in the second 
place: Arg. omn. epist.; Rom. ii 26; iii 19, 24; vi 19; viii 3, 13; xii 8; 
XV 30; 1 Cor. ii 9; x 27; xi 27, 31; xiii 2, 3; xiiii 19; xv 28; 

2 Cor. xi 13; Eph. ii 2; vi 5; Col. i 23; 1 Tim. ii 8; 2 Tim. iiii 7. 

satio (verb): Rom. xv 24; 1 Tim. vi 9. 

scilicet is an extremely favourite particle, uidelicet being very 
rare; the former occurs: Rom. vi 12; vii 17, 23; viiii 8; xi 26 
xii 18; xiiii 20; xvi 15; 1 Cor. i 2, 30; ii 8; iii 21; vii 3; viiii 12 
xi 8, 29; XV 8, 24, 25, 28; 2 Cor. i 17; ii 13; iii 5, 9; Gal. i 1 
iii 21; iiii 24, 28; Eph. iii 18; iiii 22, 24; Phil, ii 6, 11; iii 3, 15 
1 Thess. ii 12; 2 Thess. ii 3; iii 11 ; Col. i 10, 23; iii 15, 16; iiii 6 
1 Tim. ii 14; iii 8; iiii 2; vi 13; 2 Tim. i 6, 9, 10; iii 2; iiii 5 
Tit. i 2, 3, 7, 14, etc. 

sector (usually metaphorically, with an accusative indicating 

1 See Study of Ambrosiaster pp. 146 f . : occurs in Tert. pat. 8; cam. resurr. 8; 
Hier. epist. 17, 1; Aug. cat. rud. 4 §7; Marcellin. ap. Aug. e^*- 136, 1. 

2 Add to the passages in Study of Ainbrosiaster p. 146, n. 2, Ambr. expos, ps. 
cxviii 14, 28, 1; 15, 15, 1. 


the object of pursuit or desire): Rom. x 2; 1 Cor. vii 38; xii 31; 
xiiii 12, 39; 2 Cor. xi 12; Phil, iii 15; Col. iii 3; 1 Tim. iii 8; Tit. i 7. 
secundum (preposition with accusative, far commoner than 
iuxta'^, which is very rare in this author): (a) Avith nouns: historiavi 
Rom. X 8; 1 Cor. i 2; legem (Rom. ii 29); 1 Cor. vii 39; 2 Cor. 
xiii 1; Col. iii 18; with other nouns: Rom. iii 24; vii 3; viii 26 
xiii 12; 1 Cor. i 23; iii 10, (23); vi 20; vii 3, 16; xi 3 bis, 12, 25 
xii 2, 13; xv 24 ter, 2 Cor. i 17; v 19 ter; vii 9; viiii 7; xi 6 
Gal. i 4; iiii 23; Eph. vi 1; Phil, i 20; ii 5 qiiater, 22; iii 5 
2 Thess. ii 3; Col. i 15 bis; ii 7; 2 Tim. i 13; Tit. i 8; iii 5; Philem 
16 etc.; (6) with pronouns: quod; Rom. i 8; ii 13, 16; iiii 3, 15 

1 Cor. xi 18; Phil, ii 5; Col. ii 23; 1 Tim. i 18; 2 Tim. ii 8, 24 
illud 1 Cor. V 3; Gal. iii 4; Phil, iiii 5; hoc Rom. ii 16; Gal. i 1 
eos Rom. viiii 21, 26; ilium 2 Cor. v 8; nos 2 Cor. v 8. 

sensus : especially in the phrase hoc habet sensus, but also in 
other phrases; e.g. Rom. viiii 28 historia hoc habet sensus; 1 Cor. 
i 19 hoc habet sensus humanus, ut...; 1 Cor. iii 1 hoc solum ha- 
bere sensum domini, quod...; 2 Cor. i 11; iiii 15 hoc habet sensus, 
ut; in other connexions, e.g. Rom. viiii 29; 1 Cor. i 19; 2 Cor. 
viiii 5; proprio sensu, 1 Cor. viiii 8. In the plural, 1 Cor. vi 1. 

similo intr. 'I am like': Col. iii 10, 15. (Ronsch, Semas. Beit7\ 
III p. 76.) 

non solum^sed etiam is almost invariable for 'not only' — 
'but also.' Once or twice we find non solum — sed et. 

solum modo, 2 Cor. vii 12; 1 Tim. ii 5, 8, besides the classical 
tantu'm modo, which occurs e.g. Rom. xi 3; xv 24; Phil, ii 5. 

subauditur (never suhintellegitur): Rom. i 13; iii 27; v 18; 
(subaudias Rom. v 13); vii 4; x 19; 1 Cor. xi 17; 1 Tim. v 10. 

subdolus, subdole: Rom. xvi 16 non ficto uel subdolo oscido; 

2 Cor. i 12 non subdole uel astute docentes; 2 Cor. iiii 3 non- 
credentes nel subdoli; 2 Cor. xi 13 subdoli] subdolus est, qui 
aliud fingit, cum. aliud agit. 

su(b)scriptio (=' postscript') 2: 1 Cor. xvi 23; Gal. vi 18. 

superfluus, superflue (never superfluo): adj. 1 Cor. viiii 20; 
xiv 7; XV 2; Gal. v 26; Tit. i 11; iii 9; adv. 1 Cor. vii 18; xv 2; 
Gal. ii 21 ; 1 Tim. vi 10. 

taliter: Rom. iii 6; viiii 4; 2Cor.ii3; Gal.arg.; iiii 15,24; Eph.vi9. 

^ It occurs e.g. 2 Cor. V 19; vi2; Col. i 15. ^ The verb sMscn6o, 2 Thess. iii 18. 


tam...quam: Rom. ii 12; iii 20; iiii 24; xv 25, 30; 1 Cor. i 5; 
vii 14; viiii 21; xv 43; 2 Cor. ii 16, 17; vi 3; viii 23; viiii 9; 
X 5, 7; xi 13; Gal. iiii 26; v 26; Eph. ii 14; iiii 7; v 22; Phil, 
iiii 21; 1 Thess. i 1; iiii 15; Col. i 18; ii 23; 1 Tim. i 4; iii 2, 6; 
iiii 5; v 4, 7; 2 Tim. iii 8. 

tango, exactly as in Ambrosiaster', whether the word bears the 
mild meaning 'allude to,' or the stronger meaning 'attack': Rom. 
i 16 simul et illos haereticos tangit; Rom. xv 5 ut ipsos tangat, 
qui talia exercehant; Rom. xv 20 hie et pseudo-apostolos tangit; 
1 Cor. i 12 sub nomine apostolorum pseudo-apostolos tangit; 2 Cor. 
ii 17 pseudo-apostolos tangit; 1 Tim. v 6 nostr^i temporis uiduas 
tangit; 1 Tim. vi 1*1 principcdem eorum tetigit morhuni. (Cf. taxo, 

tantum ut : Rom. xiiii 9 ; 1 Cor. vii 39 ; 2 Cor. xiii 7 ; 1 Thess. 
V 22-. tantum ne: 1 Cor. vi 11; x 25; Gal. v 13^ 

taxo, taxatio: Rom. i 16 hoc ad taxationem suptiliter pei'tinet 
paganorum; Col. ii 11 hie iam pseudo-apostolos taxat. (Cf. tango, 

tempus, in the abl. tempore, contrasted with another category : 
cf. Rom. ii 11 'primum' credulitatis tempore, non honore; 1 Cor. 
XV 9 'minimus' tempore, non labore (contrast 1 Cor. xv 23 tem- 
poris uel honoris); Eph. iii 8 'minimum' tempore, non labore; 
Col. i 15 'primogenitus/...non tempore, sed honored 

terrenus^ (greatly preferred to terrestris, and opposite to cae- 
lestis): Rom. vi 6; 2 Cor. vi 10; viii 2; Eph. i 3; ii 3; Phil, ii 1; 
1 Thess. V 12; Col. i 20; iii 1; 1 Tim. iii 8; 2 Tim. ii 10; 
Tit. i 7. 

tolerantia (a favourite word of Augustine): 2 Cor. i 6; Phil, 
i 7; iiii 9; 1 Thess. i 3, 5; Col. iiii 18. 

tolero: Rom. viii 18, 36; (xv 4); 1 Cor. viiii 12; Gal. iii 4; 
Tit. iii 3. 

tollo c. ace. et dat. Rom. ii 14; xi 2, 33; xiii 10 (tulerit); xv 9; 

1 Cor. xiii 3; Gal. ii 14; Phil, iii 13; 1 Tim. iiii 2. 

1 Cf. study, p. 143. 

2 Also in the scripture of 2 Cor. iiii 13 ; Gal. vi 12. 

3 And the scripture of Gal. v 13. 

^ For the type of note, cf. Rom. xvi 13 aetate, non partu ; cf. also 1 Cor. vii 3 ; 

2 Cor. vi 17; Eph. v 31. 

^ See also under nobilitas above. 
s. p. 8 


tracto (a) c. obj.: Eph. v 4; (b) c. de: Rom. xiii 8; 1 Cor. vi 12; 
xiii 3; xv 24; 2 Cor. viiii 2; 1 Tim. ii 15. 

transitorie (a great favourite with Augustine) : 2 Cor. viiii 5 ; 
1 Thess. ii 2. 

triplex (cf. duplex above) 2 Cor. vii 7 ; 1 Thess. iii 9. 

tunc apodotic: (a) tunc — si Rom. iii 24; xii 11; xv 32; 1 Cor. 
viiii 17; 2 Cor. ii 3; vii 1; Eph. i 2; Phil, iiii 9; Col. i 23; 1 Tim. 
iiii 15; (6) tunc.quando (quandc.tunc): 2 Cor. viii 2; Eph. 
i23; 1 Thess. v 3; Tit. i 3. 

typus: Rom. ii 26; 1 Cor. x 2; Gal. iiii 24. 

ualeo c. infin. Rom. viii 36, 39; x 8; 1 Cor, i 24; 2 Cor. xiii 3; 
Gal. ii 17; Eph. iii 4, 18; vi 17; Phil, i 20; 1 Tim. i 13; 2 Tim. 
iiii 5; (otherwise used, 2 Cor. x 4). 

uere. Pelagius very much aftects the use of this word: Rom. 
XV 1, 6, 12; 1 Cor. i 7, 10, 21; iiii 5; vii 15; viii 2; x 6; xi 10; 
xii 23; xv 2, 14; 2 Cor. i 19; iii 3, 6; vi 7; xi 1, 33; Gal. arg.; 
iii 1; iiii 19; Eph, ii 7; iiii 21 qiiater; Phil, iii 10; iiii 21; 1 Thess. 
i 6; 2 Tim. ii 9. 

uindico, especially with in and the ablative^; Rom. iii 4 his, 5; 
xi 22; 2 Cor. i 23; xiii 3, as well as in the classical construction 
in and the accusative; Rom. vi 16; 2 Cor. xiii 4; with cum and 
abl., 2 Cor. ii 10; and absol., 2 Cor. xii 19, 20; xiii 7. 

usque ^ dum 2 Cor. iii 13; Gal. iii 19; usque quo Rom. xi 8; 
1 Cor. iiii 19. 

ut, quo modo..., ita (et): this collocation: Rom. i 17; vi 4; 

1 Cor. xvi 24; Eph. i 3. He uses also ut, sicut. . . , ita et 2 Cor. viii 8. 

uterque: the plur. utrique employed less exactly in the sense 
of iderque, a usage, which, according to one family of MSS; is 
found once even in Caesar^ The examples here are: — Rom. i 29; 
v 4; 1 Cor. xv 45 ; xvi 22; 2 Cor. vii 1 ; viiii 2; Gal. v 14; Phil, i 23. 
But though the author makes these slips*, he knows the correct 
uses of the singular and plural quite well: e.g. sing. Rom. x 5; 
xiii 9; 1 Cor. viii 6 his; 2 Cor. iiii 18; vi 8; vii 11; xii 2; Gal. 
iii 13; Eph. ii 14; v 24; Phil, ii 13; iii 7; 1 Tim, vi 16; plvr. 

2 Cor. ii 15; viii 14; Gal. v 19. 

1 So uindicta in 2 Cor. ii 1-5. lu 2 Cor. iiii 4 uiinlicare lias also an object. 

2 See quoad usque above. 

* So quique = quisque 1 Tim. iii 1, ■^ Cf. Aug. ep. 126, 11. 


utor libertate: cf. Rom. xiii 1 ita debere libertate Christiana 
uti, ut...; Rom. xiiii 16 non ea (libertate) ita debemus uti, ut...; 
1 Cor. viiii 19 cum possim uti libertate mea; 1 Cor. x 29 ita utor 
libertate mea, ut; 2 Cor. i 24 ne ipsa Mhertate us\ fueritis in peius. 

Supplementary Note 

While finally revising this chapter I noticed some other examples that 
might have been adduced in support of my argument. I add them here with- 
out references : 

The following words or phrases introduce notes : Hoc {totum) agit ut, 
Increpat, Modo, Plus, Praeuenit. 

A second quotation from the same author or book of scripture is intro- 
duced by et iterum. 

The genitive singular of present participles like confirmantis is rather 
frequently employed. 

It is not infrequently mentioned that some quality belongs et mentis et 

The use of the following words, senses or constructions is also character- 
istic : compello, confirmo, copulo, definio (especially definiuit), deprecor ( =pre- 
cor), dominor (c. dat.), duco ('I think'), expeto,fons (metaph.), gida (metaph.), 
inhaereo (c. dat.), permaneo, praecipio (c. ace. et inf.), putor (c. gen., 'I am 
thought guilty of), suffragor. 

The Authenticity of the Prologue and Arguments 

That the prologue and arguments are by the same author as the expositions 
themselves, will not be questioned by any one who has read through this 
chapter carefully. Those who de.sire to test the matter by itself for them- 
.selves, would perhaps find it simplest, in reading the prologue and arguments, 
to select any striking words or expressions, and then consult the preceding 
lists. The authenticity of prologue, arguments and expositions hangs to- 





In this chapter an attempt will be made to describe the textual 
character of the Bible habitually used by Pelagius. An index of 
passages quoted shows his intimate acquaintance with scripture in 
every part; very few books of the Bible are left unquoted or un- 
referred to. It is true that from most parts of scripture the quotations 
are short, but even these may be instructive. To set over against 
this general brevity of quotation, also, we have received through 
him a complete, or almost complete, text of the Epistles of St Paul 
in Latin, and if we can only fix this text amidst the varying testi- 
mony of the MSS at our disposal, it will be no small gain to have 
in front of us what is substantially a Latin codex of the Corpus 
Paulinum, belonging to the latter part of the fourth, or the very 
'beginning of the fifth century. Except for the still earlier codex 
which can be reconstructed fi:om the Ambrosiaster commentary, 
we have nothing to compare with this in age till we come down to 
the femous Codex Fuldensis of Bp Victor of Capua, written in the 
middle of the sixth century. 

The proper method to pursue in studying Latin Biblical texts, 
is to examine the extent of their divergence from the Vulgate. 
At first sight this may seem an easy matter, and so it is if by the 
Vulgate we mean the Sixto-Clementine Vulgate. But the labours 
of scholars during the past two centuries have made it clear that 
we can improve upon the Sixto-Clementine edition, though it is 
by no means a careless piece of work even from the textual point 
of view. If, however, we mean by the Vulgate the revision of the 
Old-Latin New Testament and the translation of the Old Testa- 
ment as they respectively left the hands of St Jerome, we are face 
to face with considerable difficulties. 

It is no very hard matter to identify a Vulgate Old Testanfent 
or the Vulgate Gospels. In the former case the fact that it was 


translated directly from the Hebrew marks it off from all earlier 
translations, quite apart from the presence of the prefaces by 
Jerome himself with which various parts are equipped; in the 
latter case the presence of the famous Nouum opus facere me cogis 
preface is a sufficient sign of what is Vulgate, what is not. But 
when we come to the Pauline epistles or other parts of the New 
Testament than the Gospels, how are we to say what is Vulgate 
and what is not? St Jerome thrice says that he revised the 
(whole of the) New Testament^; but if he did, his revision appears 
to have been perfunctory, and he evidently did not take the 
trouble to write prefaces to the Acts, Apocalypse, Catholic Epistles, 
or Pauline Epistles. And if he did indeed revise the whole, what 
proof have we that the text in the Sixto-Clementine edition, or 
even in the Wordsworth-White edition of Acts and Romans, has 
ever really passed through the hands of Jerome ? May not his 
revision have perished altogether? He does not seem to have felt 
much interest in this New Testament textual work, or to have 
taken any steps to secure that it should be perpetuated. It began 
under compulsion of Pope Damasus, and Jerome's way of quoting 
the New Testament subsequent to the supposed date of his re- 
vision, appears absolutely to ignore his own work. The matter is 
yet further complicated by the fact that nearly every codex of the 
Pauline Epistles in the Vulgate text is provided with a prologue 
which is the work not of Jerome, but of Pelagius himself! 

As some basis of investigation is absolutely necessary, it will 
be assumed in this book that the text published in the Editio 
Minor of Wordsworth and White's Vulgate New Testament^ is 
really what we understand by the Vulgate, in the purest form at 
present attainable. The earliest fixed date at which the whole of 
the Vulgate New Testament is known to have existed, is the first 
half of the sixth century, the age of Cassiodorusl Cassiodorus 
possessed a complete Old-Latin Bible and a complete Vulgate 
Bible. It was not till the ninth century, however, that the 

1 See De Vir. Inl. 135; epist. 71 § 5 (C.S.E.L. lv p. 6 L 10); epist. 112 § 20 
(C.S.E.L. Lv p. 391 11. 3—4). 

2 Oxonii 1911 (published Jan. 17, 1912). 

3 The genuine Augustinian Speculum provides a Vulgate text, and I believe that 
this work may be in the state Augustine left it, but some scholars still think that, 
in the form in which Augustine issued it, it was still partly Old-Latin. 


Vulgate gained complete ascendancy. Down to that date pure 
Old-Latin, pure Vulgate, and mixed texts were in circulation. 
Wordsworth and White's text may be taken to represent the form 
which the Vulgate New Testament text had in the sixth century. 

The presence of the Pelagian prologue in nearly all Vulgate MSS 
of the Epistles of St Paul is a sufficient proof that the Vulgate of 
these Epistles and the Pelagian commentary have in part a common 
history. How far the fact of that common history is due to the 
attribution of the commentary to Jerome in certain of its manu- 
scripts, may be partially decided by the investigation that is to 
follow. But whatever may be the verdict of scholars as to the cor- 
rectness of my view on the question of the text used by Pelagius 
himself, there will be furnished in these volumes such a body of 
entirely fresh evidence on the history of the Latin text of the Epistles 
of St Paul in the period between a.d. 400 and 800 as no other in- 
vestigator has ever been privileged to publish, and I shall be well 
content if other scholars, using the evidence with which I provide 
them, are able to attain to truths hidden from myself in spite of all 
my strenuous endeavours to grasp them. 

It is a working hypothesis^ with investigators of the Biblical 
texts employed by ancient Christian writers, that if certain manu- 
scripts of these authors' works offer the Biblical quotations in a text 
of normal character, such as the Constantinopolitan text in the case 
of Greek quotations and the Vulgate text in the case of Latin quo- 
tations, while other manuscripts give the quotations in an earlier 
form of text, the latter manuscripts are rather to be followed by the 
editor. I doubt if any one will be found to maintain the contrary: 
for it is hardly to be conceived that any scribe would put himself 
to the trouble of altering Biblical quotations, which were before his 
eyes in a normal type of text, back into an earlier type, for which 
for some reason he had a predilection. Similarly, if there are two 
manuscripts or two families of manuscripts, one of which more nearly 
approaches the normal in the character of its Biblical quotations 
than the other, the latter is the manuscript or family to be followed 
by the editor in constituting the text. 

^ Cf. the writer's Text and Canon of the Neic Testament (London, 1913) pp. 17, 
85 f., as well as many other works. 

iv] the biblical texts used by pelagius 119 

§ 1. The Text of the Pauline Epistles 

The investigation detailed in the second chapter resulted in the 
conclusion that two manuscripts alone present the Pelagian com- 
mentary in its original form, A Codex Augiensis cxix (saec. viii — 
IX) at Karlsruhe and B Balliol College MS 157 (saec. xv med.) at 
Oxford. We must start with the assumption that they also preserve 
the Pauline text in the form approved by Pelagius. We shall see, 
however, that these two manuscripts are far from showing the same 
exact agreement in Biblical text that they show in the part which 
is exposition. They share many non- Vulgate readings, but each on 
occasion supports the Vulgate reading against the other, and if we 
follow the rules stated above, we shall by their joint aid have to 
constitute a text that is as far removed from the Vulgate as possible. 
In other words, wherever one of these manuscripts reads with the 
Vulgate, the other against, we must at that point follow the latter, 
whichever of the two it may be. We shall of course have the weight 
of other manuscripts to help us, and may sometimes find that neither 
A nor B is right, but that we must follow some other MS whose 
text explains the aberrations of both. But before proceeding to the 
investigation proper, we can get some preliminary light on our 
problem from other considerations. 

We shall discover in the course of this chapter that the quota- 
tions from every other part of the Bible than the Pauline Epistles 
are made according to an Old -Latin text. Nay more; the quotations 
from the Pauline Epistles themselves made in the course of the notes, 
apart from the lemmata altogether, and therefore much less exposed 
to alteration on the part of scribes, are also in the vast majority of 
cases made from an Old-Latin text, and in the few cases where they 
are not, there was perhaps no room for variation. I am of course 
not blind to the fact that a writer might employ more than one type 
of text, might, in fact, employ one type consistently for lemmata, 
and yet refer consciously or uncon.sciously to a different type in his 
notes. It is probable, for example, that Origen did so in some of 
his commentaries. It was also quite possible for a writer to employ 
a Vulgate text for one section of scripture, and a non- Vulgate text 
for another. Augustine, for instance, after about a.d. 400, uses the 
Vulgate of the Gospels when he seeks to quote diserte from the 


Gospels, while he employs an Old-Latin text for the rest of the 
New Testament, and even for the Gospels (in this case akin to eY 
when he is quoting from memory. Jerome, also, used his own Vulgate 
in commenting, for example, on Jeremiah, but for a number of 
Biblical books he consistently employs an Old-Latin text. The 
possibility that Pelagius used the Vulgate, or something very like 
it, for the lemmata, cannot at this stage be denied. Yet the pre- 
sumption appears to be against this. 

The parallel case of Ambrosiaster is instructive. No one doubts 
that this author employed an Old-Latin text for the lemmata of 
the Epistles, and yet in certain MSS, as Father Brewer informs me, 
the Vulgate text has been substituted throughout in the lemmata. 
The analogy of this case strongly favours the view that the great 
value of both commentaries, Ambrosiaster and Pelagius, created the 
demand for editions of these in which the Vulgate, with its growing 
influence and importance, was substituted for the text which the 
author employed. There can be no doubt that both Ambrosiaster 
and Pelagius arranged their work consistently throughout in the 
order: — (a) clause or clauses of text, (6) comment thereon'l At a 
later date, in the case of Pelagius at least, it became a practice to 
copy the whole commentary in the form of interlinear glosses, into 
a text of the Epistles already completely written, the Pelagian 
lemmata being either partially or entirely ignored. It is time, 
however, to pass from general considerations to a detailed exami- 
nation of the question. And this we shall do under four heads: 
(a) occasional references to variae lectiones by Pelagius himself; 
(6) quotations from the Epistles made in the body of the notes; 
(c) the light thrown by the comments on the character of the text 
which lay before the author; {d) the testimony of the Vatican 
fragments and the interpolation in Ambrosiaster. 

(a) Occasional referencefi to variae lectiones 

by Pelagius himself 

Rom. xii 13. He clearly prefers necessitatibus, though he refers 

to memoriis as a variant. Now necessitatibus is the Vulgate reading, 

with almost no Old-Latin support; me?)iorm is the Old-Latin reading, 

1 Cf. J.T.S. vol. XII (1910—11) pp. 154 f. 

2 This is proved for Pelagius in chap, ii p. 50. 


supported by the one, but very important Vulgate codex, Amiatinus, 
which is however not backed up by Cassiodorus here. 

1 Cor. X 22. He comments on An aemulamur (adulaniw) domi- 
num? He mentions a variant occurring in other manuscripts: Ipsi 
me zelauerunt in non deo. As no other Latin authority is known for 
this latter reading at this point, I think it probable that Pelagius 
is referring to the Old-Latin text of Deut. xxxii 21, from which verse 
this Pauline extract comes (avrol irape^TjXcoadv fxe eV ov 6ew). 

2 Thess. ii 3. Here discessio was in his lemma, but he notes the 
variant refiiga. Discessio is the Vulgate reading, definitely approved 
by Jerome himself {epist 119 §7 = C.S.E.L. LV p. 455 1. 12; also 
epist. 121 § 11 = C.S.E.L. LVI p. 53 1. 12): refaga is the most preva- 
lent of various Old-Latin renderings of cnrocnaaLa^. 

Col. iii. 15. While reading grati with all other known authori- 
ties, he says that some copies have gratia. To the best of my 
knowledge, no other authority for gratia has turned up. 

Such are all the definite references to various readings in 
Pelagius. So far as they go, they are not unfavourable to the view 
that the Vulgate was the basis of his comments. 

(6) Quotations from the Epistles made in the body of the notes 
As is hinted above, these appear to come from an Old-Latin 
source, but they must be studied in some detail. The method here 
followed is to compare each quotation with the Vulgate form, and 
to set down the word or words which show a real difference from 
the Vulgate. Then we search for other authorities in support of 
those readings. These where they are forthcoming, are arranged in 
three classes, first, manuscripts which rank as Vulgate MSS, de- 
scribed by the capital letters used by Wordsworth and White to 
indicate them; second, Old-Latin MSS, described by the usual 
minuscule letters; third, the names of Latin writers who quote 
according to that form, the names of these writers being indicated 
by the same abbreviations as are employed in the Novum Testa- 
mentmn Graece (Oxonii, 1910, 1913, 1918) pp. xvii — xxii. An 
apology is due for any defects which may be noted in the enumera- 
tion of these authorities. The whole body of the authorities not 

1 See now the new fragments from Africa, published by Monsieur H. Omont, 
Comptes rendus des seances de I'Acad. des Inscr. et Belles-Lettres, 1918, pp. 241 — 250; 
cf. P. Alfaric in Rev. d. Hist, et de Litt. relig. vi (1920) pp. 62—98. 




being accessible to me, it is proper to say that I have used 
Wordsworth and White's Romans; Nestle's Novum Testamentum. 
Latine (Stuttgart, 1906); Wordsworth and White's Novum Testa- 
mentum Latine: editio minor; Dr Gwynn's Liher Ardmachanus; 
Buchanan's Sacred Latin Texts: Nos 1 aiid 2 (London; 1912, 1914); 
Tischendorf's Codex Claromontanus; Reichardt's Codex Boerneri- 
anus (Leipzig, 1909); Weihrich's Speculum (m); and Sabatier's 
great compilation, controlled in part by modern critical editions in 
the Vienna series, Ronsch's Das Keiie Testament Tertullians, Hans 
von Soden's Das Lateinische Neue Testament in Afrika zur Zeit 
Cyprians, personal study of Victorinus's text in Migne P.L. viii. The 
Study of Ambrosiaster, Swete's Theodore of Mopsuestia, and the 
Toledan Lectionary (Liber Comicus^) etc. Where no authority is 
added, it means that I have found none. 
Rom. ii 1 qui {pro in quo), 
alium iudicas. 
D (m) Aug. 
v 20 peccatum {alt). 

Iren.'** Orig.'^* 
vi 8 om. lesu. 

Tert. Ambr. Orig.i^*- 
vii 12 c\j iustum et sanctum, 
viii 3 legi BDKUVWZ'' vg^i^'"- 

Qj.jg_lat. Aug. Vigil, 
viiii 6 qui] + sunt. 

Orig. Tycon. Ambst. Aug. 
X 12 deus. 

xi 23 inseruntur (?) Hil.'=°'^- 
xiii 10 caritas {\) {pro alt. dilectio) MZ^*'*'^- (m) Ambr. Aug. 
xvi 17 obseruate (d*gm). 
„ 18 deseruiunt^. 
1 Cor. i 23, 24 om. quidem Ambr. \ Leo. 
uero {alt.) t. 

om. ludaeis atque Graecis. 
^ Published by Dom Morin as Anecdota Maredsolana vol. i (Mareds. 1893). 
2 A favourite word of Pelagius: see chap, iii s.v. 

Ambst. Tycon. Hier. Aug. 




1 Cor. ii 16 

iii 3 

V 4 

vii 7 


viiii 24 



xiii 2 

xiiii 14 




Cor. V 









domini (pro Christi) D g r Ambst. 

homines (pro carnales). 

cum (pro et). 

om. cum uirtute domini lesu. 

et alt] + in. 

CO esse omnes homines. 

om. ipsum. 

enim (pro alt. autem) ^ D Ambst. Ambr. Vigil. 

om. quod | dg Tert. Ambst. (Greg. Illib.) Ambr. 

Aug. (Vigil.), 
ut] + omnes ADO Ps. Cypr, Ambr, 
Graecis d g Hil. Ambst. Aug. 
csj omnibus per omnia DO. 


Tert. Cypr. Pacian. Ambst. Hier. Aug. 
ne (pro ut non) d Ambst. | Aug. Gaudent. 
habeam {pro tert. habuero) DO. 

d g m t. 

Cypr. Opt. Pacian. Ambst. Ambr. Aug. 
mihi prodest (pi'o sum) D Ambr. 
orauero f D (d uacat) g Ambst. Aug. 
benedicas (d uacat). 
occulta] + etiam vg^'^- d Ambr. 
tunc (pro ita) D d Ambst. 
c\3 deus uere. 
CO est in nobis Ambr. 

per (pro propter) D^Z'^vg^'^™ d Ambr. Aug. 
terreni] + ita Ambst. h, . 
om. et Hil. Instant, 
illi (p7-o ei). 
castos d gl r Ambst. 
sustinete (pro utinam sustineretis). 
zelo (zelans) g| Ambst. Hier. Rufin. Aug. 
zelo (noun) g Ambst. Hier. Rufin. Aug. 

Christi DF^OZ'^o"- g Ambst. (?) Faustin. Aug. (non 
Gwynn's resolution of the symbol seems wrocg here. 


Gal. ii 4 cva nostram explorare subintroierant (-unt) liber- 
iii 27 cvj baptizati sumus in Christo, 
V 1 1 om. adhuc (1°) D d g Victorin. Ambst. Hier. 
21 sicut] + et vg<^°'^<^ d g t Iren.'^'- Ambst. Hier. Theod.- 
vi 1 praeuentus m Hil. Pacian. ^ Hier. \ Arnob.-Iun. 
2 inuicem {pro alter alterius) D 

Tert. Pacian, Ambst. Aug. {non semper) Theod.- 
on era] + uestra D 

Tert. Pacian. Ambst. Hier. Aug. {non semp.) 
7 deridetur D 

Tert. Cypr. Victorin. Ambst. 
10 quod bonum est (pro bonum) D Cypr. Theod.- 

12 qui Aug. 

oo in came placere. 
minime (jaro non). 

mortui] + in Theod.-Mops.' 

fuistis Tert. | Ambst. Aug. Promiss. 

aliquando om. BHo, but AHj have it. Perhaps the 

omission is an accident, 
sine macula Victorin. Macrob.-Don. 
quasi {pi-o ut) g Cypr. 

diaconis D r. 

cupio Tert. Victorin. Ambr. Hier. Aug. {no7i semp.) 
CO esse cum Christo D vg*^^^"^- 
(Tert.) Victorin. Aug. Cassiod. 
ii 8 alter alterutrum maiorem aestimantes. 
cf. alterutrum existimantes maiores D 

inuice m existi man tes su periore s (uel maiores) g. 
7, 8 CO exinaniuit se ipsum. 










. i 



Phil, ii 8 se (pro semet) dgm t|. 

Cypr. (Nouat. Hil. Lucif. Foebad.) Victorin. 
Faustin. (Ambr.) Aug. Theod.-Mops.'** 
iii 12 si] + quo modo vg*='*^™- Victorin. Macrob.-Don. Aug. 
17 GOgnoscite cf. et cognoscite et seruate D (a doublet) 

21 transformabit (-uit) g| Cypr. Hil. 
con forme D. 

Hil. (non semper) Ambst. Iren.'**^- Ambr. Hier. 
Aug. Promiss. Fulg. Theod.-Mops.^^*- 

1 Thess. ii 5, 6 <n> fuimus in sermone adulationis aliquando. 

of. Gildas. 
nee {pro alt. neque). 
14 et uos {pro uos enim). 

CO fratres imitatores facti estis D Ambst. 

am. in Christo lesu. 

eadem enim {pro quia enim). 

am. et. 

quae {pro sicut). 

illi {pro ipsi). 

2 Thess. i 2 om. nostro d. 

6 CV3 his qui uos tribulant tribulationem. 

Vigil. ^ (cf. d Iren.'**^- \ eis qui tribulant uos tribu- 
lationem, g eis tribulantibus uos tribulationem, 
Theod.-Mops.'*'^- his qui tribulant uos retribu- 
ii 11 mittet CDGHOVZ^ vg'^i^'^- 

Cypr. Iren.^^' \ Ambst. Hier. Aug. Promiss. 
spiritum {pro operationem). 
12 quia (pro qui) (cf O quod). 

Col. ii 5 corpus (|) {uacat g). 
om. sum {uacat g). 

praesens {pro uobiscum sum) {uacat g). 
iii 3 absconsa d Hil. 
apud deum. 


Col. iii 11 ludaeus et Graecus D Macrob.-Don. Faust. Ambr. 
Aug. (cf. d g Ambst. Hier. Grecus et ludaeus). 
iiii 17 om. in domino m. 

1 Tim. i 9 data {pro posita) Greg.-Illib. f . 

2 Tim. ii 12 sustinemus AHOVZ. 


Ambst. Ambr. | Theod.-Mops.'*'- 

(cf. toleramus D Cypr., toleremus d, sufferimus 


Tit. i 16 cvj deum confitentur se nosse dg^. 

(Ambst. Theod.-Mops.i** ) 

iii 8 credentes {pro qui credunt). 

om. deo. 

We may fairly conclude that passages in the above list where 
no other authority for the Pelagian readings can be adduced, are 
quotations from memory or paraphrases, and may therefore be left 
out of account. An examination of the others leaves no room for 
doubt that there is a special kinship between the copy of the Epistles 
used by Pelagius and D {Book of Armagh), in cases where the 
latter MS diverges from the Vulgate. A reference to Pelagius's note 
on the passage will show that the quotation from Phil, iii 17 is 
particularly significant. In the remaining passages above where 
Pelagius's quotation does not go with D, it goes with some copy of 
the Old-Latin; particularly of the European Old-Latin. It is often 
found in company with d and g, and where d and g diverge, rather 
oftener with g than with d. Now D is a composite text, partly 
Old-Latin and partly Vulgate ^ It would seem therefore that the 
copy in Pelagius's hands was an Old-Latin text, representing through- 
out in absolute purity the particular Old-Latin element which only 
partially survives in D. We shall see later that this conclusion is 
confirmed. Dr Gwynn appears to regard the basis of D as Vulgate, 
and the Old-Latin elements as intrusions. May it not rather be 
that, as the basis of the Vulgate itself is Old-Latin, the basis of D 
is also Old-Latin, and the Vulgate elements are superimposed on 
its original fi^om a copy of the Vulgate ? 

1 See S. Berger, Histoire de la Vulgate (Paris, 1893) pp. 31 ff. 


(c) The light throivn hy the comynents on the character 
of the text which lay before the author 

The conclusions of such a paragraph as this must be received 
with some reserve. Pelagius, as we saw above\ was acquainted with 
various readings, and it is not at all impossible that he may have 
consciously or unconsciously employed in a note a synonym of 
some word in the text, which synonym may or may not have really 
been in the biblical text on which he intended to comment. Yet 
some of the results of this paragraph seem indisputable. 

Rom. xi 11. The comment on this verse, Usque adeo illos dilexit, 
etc. seems to indicate that he read in the text dilecto, not delicto. 
It may be remarked that there is no reference to a delictum in the 
note. Dilecto happens to be the reading of the first hand of Amia- 
tinus. If my inference be unjustified, possibly Pelagius connected 
delictum etymologically with diligo: no one who knows anything 
of ancient etymology will doubt such a possibility. 

Rom. xi 17. The note, radicis patrum, pingvedinis Christi 
makes it almost certain that these words occurred in the same order 
in the lemma. As there is no authority for the order pinguedinis... 
radicis in the lemma except A, this is clearly a freakish error of A. 

Rom. xi 32. The note has quos inuenit IN INCREDVLITATE. 
This suggests that we ought to read the same in the lemma, with 
a number of Vulgate MSS and Old-Latin authorities, but we cannot 
be certain on the point. 

Rom. xii 2. The note is QVOD bonvm sit et melius et optimum. 
This suggests that we ought to read in the text Quod bonum [est] et 
beneplacitum et [quod] perfectum [est] with DO* d* g gue m Ambst. 
Qj,jgiat. Aug., and in fact our AB also. 

Rom. xii 17. The note id nan deo sed solis uominibys placere 
desideres favours the idea that we ought to leave out omnibus 
in the lemma, as is done by A (def B) in combination with 
DT d g gue m t Ps.-Cypr. Lucif and some Greek authorities. 

Rom. xiii 9. The note recapitvlatvr omnis iustitia in proxivii 
dilectione, tempts one to read recapitulatur in the lemma, with g 
Hier. Orig.^**- Aug., but no Pelagian MS (def B). It is possible that 
Pelagius took the word from Origen-Rufinus, which Dr Smith has 
proved that he used^ 

1 Section (a). 

2 See J.T.S. vol. xx (1918—19) pp. 127 ff. : also below, pp. 188 ff. 


Rom. xiii 12. The note arma lucis, hoc est liuninis opera, 
INDVAMVS, suggests that we ought to read in the lemma induamus 
arma lucis with DFL* d g t Cypr. Orig.'*' Aug. Gildas etc. 

Rom. XV 4. The note id per exempla patientiae et consolationis 
eorum quae scripta sunt, speremus consolationem et in praesentibus 
temptationibus etc. seems to favour A in reading in the lemma spent 
haheamus consolationis with U, the Greek B and Clem.- Alex. 

Rom. XV 16. The note serviens in euangelio; hoc est, reuocans 
ei seruos quosdam fugitiuos favours the view that the lemma should 
read ut sini seruiens etc. with ABDd*g. 

1 Cor. i 10. The note si unum sentiatis et proferatis rather 
favours the reading of the lemma thus: in eodem sensu et in eadeni 
sententia, with ACDG^O(Z) vg^i^''^ Cypr. Hil. Aug. etc. 

1 Cor. ii 3. The note neque enim alio niodo devm conuenerat 
praedicari suggests that we should read in ii 1 dei (not Christi) 
with A (om. B altogether) d g r Ambst. Ambr. Hier. Aug. Christi 
seems in fact to be a real Vulgate reading. 

1 Cor. viiii 5. The note Non dixit mvlieres 'ducendi'.., 
sed ' circumducendi^ favours the reading niidieres in the lemma, 
with A. The other Pelagian MSS go wrong in giving sororem 
mulierem (= vg), but midieres is supported by Dg Tert. Hil. 

1 Cor. X 13. The note Non uohis venit extrinsecus ista teniptaiio, 
suggests that we ought to read adprehendit in the lemma with A d. 

1 Cor. xi 10. The note Velamen signum potestatis esse de- 
clarat suggests that we ought to read potestatem in the lemma, 
and in fact AB and nearly all the other Pelagian MSS do so: 
potestatem is also read by vg including DOZ, dg Tert. In fact 
Iren.^**-. Hier. vg*'^ . Aug. seem to stand alone in reading uelamen. 

1 Cor. xi 34. The note ad iudiciuni conueniebant suggests 
that we ought to read in the lemma ad iudicium conueniatis with 
Ps. Hier. MSS (def. B), Ddg Cypr.(?)i Aug. 

1 Cor. XV 11. From the comment et nos sic praedicavimvs et 
uos similiter credidistis, there is some probability that the lemma 
should read sic praedicauimus et sic credidistis, though AB Ps.- 
Hier. read praedicamus. In favour of praedicauimus are FO vg*'^-. 
This, however, is a very doubtful case. 

1 Hartel gives L as reading ad ; my report from a study of a photograph of L 
gives no prep. ; von Soden reads in. 


1 Cor. xvi 1. The comment per singulas ECCLESIAS favours 
ecclesiis in the lemma, and not ecclesiae, as is read by B Ambst,*^"'^'^-, 
and yet we cannot be quite certain. 

2 Cor. i 4. The note on this verse, non est minus tribvlatione 
solacium, perhaps favours the reading of tribulatione in the lemma, 
though the synonym pressura is read there by B with D Ambst. 

2 Cor. i 24. The note credendo Christo stare coepistis seems 
on the whole to favour the reading of stetistis in the lemma with 
A, one or two Ps.-Hier. MSS, (A)Z and seemingly other good 
Vulgate MSS. 

2 Cor. vi 1. The note gratiam dei recipit seems to favour the 
reading of recipiatis in the lemma with vg DOZ Ambst., but that 
this view is at least doubtful is suggested by the fact that A and 
one Ps.-Hier. MS read excipiatis, which is the reading of d. 

2 Cor. vi 9 is a somewhat doubtful case. The true text of the 
comment would seem to be: ignorati a perjidis et ingratis et 
COGNITI [a] fidelihus atque iustis. This points to ignorati rather 
than ignoti in the lemma, where A has sicut qui ignorati et cogniti, 
with no authority known to me\ A's reading would appear to be 
a revision of the reading of B, ut qui ignoramur et cognosdmur, 
which is supported by D^dg (Ambst.) Aug., and is no doubt right. 

2 Cor. vi 9 (again). The note is a quotation from Prou. xvii 6, 
beginning with the words castigans castigavit me dominus, 
which suggests that castigati was in the lemma (= vg Paul.-NoL); 
but the agreement of ABDdgt Ambst. in reading temptati is 
difficult to get over, and we may explain the note by a knowledge 
of the other reading. 

2 Cor. vii 4. At first sight the note midtam mihi FIDVCIAM 
dedistis loquendi seems to favour the reading of fiducia in the 
lemma with vg, but fiducia loquendi corresponds so well to the 
libertas of B, supported as it is by Sedulius Scottus, that I have 
no doubt the latter was the Pelagian reading. 

2 Cor. viiii 13. The comment uses the word magnificantes 
twice here, which rather suggests that magnificantes was the word 
in the lemma. This supposition is confirmed by the fact that B 
reads magnificantes in the lemma with Ddg Ambst.^ 

1 But it should be noted that the first hand of t read ut qui ignorati et cogniti. 
- ignoremur actually in D. ^ See below on this rendering of Sofifw. 

S. P. 9 


Gal. ii 5, The comment shows clearly that the negative was 
absent from the text used by Pelagius, and this conclusion is con- 
firmed by the fact that the negative is absent from A d with Tert. 
Iren.'*^- Victorin. Ambst.f , as well as other authorities enumerated 
in my critical apparatus to the Greek New Testament. 

Eph. ii 17. The comment here contains the expression ad hoc 
VENIT, which suggests that we ought to read et ueniens euange- 
lizauit in the lemma, with vg DOZ d g (Cypr.) Hil. Ambst. Hier. 
Aug., Theod.-Mops.^^*^- in spite of the fact that A and one family of 
Ps.-Hier. MSS omit ueniens. 

Eph. ii 22. The comment has habitacvlvm dei spiritale, 
which confirms the reading habitacnlum for the lemma, given by 
vgDOZg Ambst. Hier. Aug. Theod.-Mops.^^^-, and all the Pela- 
gian MSS\ 

Eph. iiii 24. The presence in the comment of the words 
ET sanctitate veritatis proves that these words occurred also 
in the lemma, supported as they are by A and one Ps.-Hier. MS, 
with vgOZ; in spite of the fact that the words are omitted by B 
and one family of Ps.-Hier. MSS. But the true reading in the 
lemma seems to have been et sanctitate et ueritate, supported as it 
is by the other family of Ps.-Hier. MSS, with D d g m Cypr. Hil. 
(Lucif Ambr.) Theod.-Mops.'^'- ; and lest it should be argued that 
the note confirms the form first mentioned, attention must be 
called to the fact that B reads the comment as et sanctitate et 
veritate. The omission of this phrase in the lemma of B must 
therefore be due to accident. 

Phil, i 8 has in the comment ita uos DESIDERO, which suggests 
that we ought to follow B in reading the lemma thus: queni ad 
modum desiderem, as Ambst. Sedul.-Scott.; cf D dissiderauerim, 
g^ Theod.-Mops.'**- desidero. 

Phil, iii 13. The comment has ad FUlOB. A festino : we are thus 
encouraged to read in the lemma quae sunt [in] priora, as is 
indeed read by all the Pelagian MSS. The in is present in A and 
one family of Ps.-Hier. MSS, while the others omit it. The outside 
authorities are: (a) for the in: OZ Tert. Hil. Hier. Aug. Promiss.; 
(h) for the omission of in: vgDg Ambr. In priore is read by 

ygcodd. ^^ 

^ Tabernaculum must in fact be due to the first editor of Pseudo- Jerome himself 
(Erasmus t. ix, 1516). 


Phil, iiii 15. The comment profectvs svm favours the same 
reading in the lemma with all the Pelagian MSS except B, also 
with vgDOZdg(r) Ambst. Aug. as against B's exiui. Yet the 
very uniqueness of exiui (= exii Theod.-Mops.'^*^) is tempting. 

Phil, iiii 17. The comment abundans fructus orationis est 
points to the necessity for orationem uestram in the lemma, as is 
in fact read by A and all Ps.-Hier. MSS, A^Z. But B reads uerho 
uestro with d, and this, as it gives the same sense as the other, 
may be in fact the true lemma. 

1 Thess. i 6. The note in passionihus uerbum svscipientes 
suggests that we ought to read in the lemma suscipientes uer- 
hum etc. with B^ Ambst. Theod.-Mops.'^*^- Were it not for the 
support of Ambst. and Theod.-Mops.'*'-, one might be tempted to 
regard the reading of B as an accidental assimilation to the pre- 
ceding suscipientes. 

1 Thess. i 9. The note has deum vervm et vivvm in A and 
one family of Ps.-Hier. MSS, deum vivvm et vervm in B and the 
other family of Ps.-Hier. MSS. But A and the same family of Ps.- 
Hier. MSS as agreed with it before, give uero et uiuo in the 
lemma, while B and its related Ps.-Hier. family of MSS give uiuo 
et uero in the lemma. Each MS and family is therefore consistent 
with its lemma. Outside we have no support for uei'o et uiuo 
except the Verona MS (Paris 653), but this seems enough to 
justify the inversion of the familiar order. 

1 Thess. iii 2. The note has adivtores pro discipulis etiam 
uulgo dicuntur. This suggests that we ought to read in the lemma 
et adiutorem dei in euangelio Ghristi, with AB, one family of Ps.- 
Hier. MSS^ dgl''. 

1 Thess. iiii 11. The note uel proprii negotii cura et labore 
manuum suggests that the word proprius should appear in the 
lemma instead of uester, and so in fact it does according to B, 
which reads ut quieti sitis et negotii proprii et operandi etc. The 
nearest form to this I know is d Theod.-Mops.'*'^- et agere j^ropria. 

1 Thess. V 23. The note reads usque in diem iudicii. We shall 

1 As corrected from suscipies. 

- The other family has the doublet adiutorem ministrum, 

* Curiously the similar word auditor is also sometimes an equivalent of dis- 
cipulus: see Mayor on Juvenal Sat. i 1. 



therefore not hesitate to read die in the lemma, with AB (one 
family of Ps.-Hier. MSS has diem) Ambr.f. 

2 Thess. iii 1. The comment runs 'currat' afluentia, 'magni- 
ficetvr' auditorum pi'ofectii. The character of this comment 
makes a strong case for magnificetur in the lemma with D^dg|: 
glorificetur is read by B with g^ Aug. Theod.-Mops.'*'- Sedul.-Scott. : 
olarijicetur by A Ps.-Hier. vg OZ Ambst. Max.-Taur. 

2 Thess. iii 6. The words sepaeant and svbtrahendvm both 
occur in the note; so that it is a nice point whether in the lemma 
we ought to read ut separetis or ut subtrahatis. The former is read 
by BD g^ Macrob.-Don., the latter by all Pelagian MSS other than 
B and by vg OZ d m Theod.-Mops.^*' etc. We shall probably be 
right in adopting the non-Vulgate reading as lemma. 

2 Thess. iii 6. The recurrence of the expression INQVIETVDO 
(on vv. 9, 10 etc.) suggests that the expression has occurred in some 
recent lemma. We find it in verse 6, where B has inquiete ambu- 
lante with (DF)dgm^ Lucif Aug. 

Col. i 8. The comment, by reading caritatem, favours cari- 
tatem in the lemma, with BD g|. Theod.-Mops.^*'^- 

Col. i 12. The note plus uos vocatio etc. shows that the 
lemma must have read qui uocauit uos with AB, one family of Ps.- 
Hier. MSS, (D)(dgm) Ambst. Vigil. 

Col. ii 5. The comment on this verse is somewhat puzzling at 
first: et svpplens id quod de-est utilitati jidei uestrae: uel litteris 
IMPLEO quod praesentia non possum. The real explanation is that 
down to uest7'ae the passage is not a comment at all, but a lemma, 
and a lemma very different from the Vulgate: cf d nt in quod 
de-est necessitatibus fidei uestrae idef. g), Ambst. et supplens id quod 
de-est utilitati Jidei uestrae in Christum, Aug. et id quod de-est 
fidei uestrae in Christo, Facund. et id quod de-est utilitatis fidei 
uestrae. Sabatier (after Latini^) acutely observes that these curious 
readings are due to crTepecofia in the Greek being miswritten or 
misread as varepTj/Mi. The supplens is clearly introduced to make 
sense. This one passage by itself would prove that Pelagius did 
not use the Vulgate as the basis of his comments. 

^ This is a case where D may be used to correct the errors of Pelagian MSS. There 
may be other instances also. See also 2 Cor. viiii 13 above, and pp. 139 f. below. 
* The true reading: Weihrich is wrong here. ^ Biblioth. (Rom. 1677) p. 143b. 


Col. iii 5, The comment, nihil prodest deo nomine et IDOLIS 
operibus desei'uire^ gives just a suggestion that idolorum should 
be read in the lemma, and this suggestion derives support from 
the fact that idolorum is read there by one family of Ps.-Hier. 
MSS, with d g Cypr. Tycon. Ambst. Hier. Aug. Theod.-Mops.^^*- 

Col. iii 21. The comment ne exasperati iracvndi fiant suggests 
that we ought to read iracundiam in the lemma, and this is in 
fact read there by A and Ps.-Hier. MSS, with FOZ g J m'"'"^^- Ambr. 

1 Tim. ii 10. The comment clearly points to castitatem in the 
lemma, and this is in fact read by AB and one family of Ps.-Hier. 
MSS, with D (pietatem et castitatem, a doublet), m r Cypr. Seuer. 
Anon. ap. S. Paulin. 

1 Tim. iii 15. The comment, in qua sola nunc ueritas stat 
FIRMATA, points on the whole to firmamentum in the lemma, and 
saves us from any temptation to read fundamentuni with one family 
of Ps.-Hier. MSS, but with no outside support. 

1 Tim. V 19. The comment on this verse suggests, what is con- 
firmed by A and one family of P.s.-Hier. MSS, that the lemma was 
without the clause nm — testihus. The clause is absent also from 
Cypr. Ambst. codd. latt. ap. Hier. 

1 Tim. vi 9. The comment, usque ad inferos demergvntvr, 
suggests that we ought to read demergunt in the lemma, and that 
is what in fact we find in B, with D m Ambst. 

1 Tim. vi 18. The word in the comment, commvnicare, suggests 
the same word in the lemma (with Theod.-Mops.^**), but can hardly 
be regarded as decisive against communicatores esse, the reading of 
B g {esse communicatores) Ambst, {libenter communicator esy. 

2 Tim. i 10. The occurrence of incorrvptio in the comment 
commends incorruptionem as the reading of the lemma, but aeterni- 
tatem is read there in B with D Ambst.'"'"- 

2 Tim. i 17. qvaesivit in the note on the whole favours the 
same word in the lemma. There A reads requisiuit, with no outside 
support. I believe it is one of the instances of retouching which we 
find in this MS. 

2 Tim. ii 4. The comment on this verse leaves it somewhat 
doubtful whether deo was present in the lemma or not. A omits 

1 For this expression, see also chap, iii p. 98. 

2 The comment renders these readings somewhat doubtfuL 


deo, with d, but I believe all other Latin evidence points to the 
nonsensical addition of deo. 

This part of our investigation leads to the same conclusion 
exactly as the preceding, that a text akin to D was that which 
lay before our author. The case of Col. ii 5, as has been pointed 
out above, is in itself decisive against use of the Vulgate by the 

(d) As the passage of time tends to obscure textual character- 
istics, it will not be out of place to ask what evidence, if any, as 
to Biblical text is furnished by the oldest Pelagian MSS at our 
disposal. These are the Vatican sixth century fragments, and the 
fifth or sixth century reconstructed interpolation from Pelagius in 
Ambrosiaster MSS. The method followed will again be to call 
attention to all variations from the Vulgate, and to append the 
other authorities which support the same readings. 

Vatican Fragments 

Rom. vii 10 in uita (in uitam B Aug. saepe). 

in morte (in mortem B Aug. saepe). 
11 occidit] + me BDW cor. uat* f 

13 super Aug. 

CO peccatum delinquens B g| Aug. 
mandatum] + legis BD. 

14 autem B0 Ambr. Orig. Aug. 
quoniam t Orig. 

viii 7 inimica est deo ABBDFKLUVW Ambst. Orig. 
est subiecta BBDFKLVWZ-^ Orig. Aug. 
Here it is very significant that out of ten readings in all where 
the Vatican fragments disagree with the Vulgate, eight should be 
shared with the Balliol MS. It is only less significant that the Book 
of Armagh should agree with four of them. 

Interpolation in Ambrosiaster^ 
1 Cor. XV 44 surget Dg (corr.) (m) Ambst. Iren. Aug. Gaud, 
(surgit Ambst.'^^). 
om. si est — spiritale ABD m. 

^ As reconstructed in chap, ii pp. 51 f3f. 


1 Cor. XV 46 om. alt. est (iiersum om. A) B d g m Tert. f Iren. Ambr. 

49 illius terreni BDZ Ambst. Ambr. 

huius caelestis B (eius c. D) (illius c. O) Ambst. Ambr. 

50 non possidebunt BDO g^ Tert. I Siric. Greg.-Illib. 

51 om. uobis AB Ambst.'=°'''^- 

0771. sed B (m) (habet Ambst.*'*®-). 

52 om. canet enim B (m) (Ambr.). 

53 incorruptionem DOZ d g m Tert. I C3rpr. Hil. | Ambst. 

Ambr. (non semp.) Aug. (non semp.). 
(incorruptelam Ambst.^**). 

55 pr. stimulus tuus B d (g) (Tert. f ) (Cypr.) (Iren. ^) 

(Hil.) (Aug.). 
alt uictoria tua (om. A) B d g (Tert. f ) (Cypr.) Iren. 
(Hil.) (Aug.). 

56 CN3 est peccatum B Aug. 
xvi 1 Nam de collectis B. 

ecclesiae B Ambst. 

2 om. ponat B Ambst.*'"^- 

3 CN3 fuero praesens (uersum om. A) B Ambst.*'°**' 
quoscumque BD d g Ambst.''"*'- Aug. 
epistulam B. 

om. in B. 

6 om. apud — hiemabo. 

ut — deducatis] nisi (+ si d) uos me duxeritis (om. A : 
deduxeritis g Ambst.''** ) d g, ut uos me duxeritis 
B, nisi me uos deducatis D. 

7 mode] nunc (om. A) B d g| Aug. 
mansurum (om. A) BD. 

8 manebo BD Ambr. 

9 enim] autem. 

et alt] sed BD Hier. 

10 quia opus (nam opus D). 

11 om. ergo AB DO d g. 

12 fratre] 4- notum uobis facio quia B d cf. (nostro D) 

n. u. f. (faciam D) quoniam DFOZ, notesco 
uobis quia g significo uobis quia Ambst. (uersum 
om. A) 


csj ilium rogaui (om. A) BD d (r) Ambst. 
ut ueniret] uenire (om. A). 
om. ei (om. A) BD d g Ambst. 
oportunum (am. A) DF^ B d g r Ambst. 
1 Cor. xvi 15 Fortunati] + et Achaici (uersum om. A) BDO g. 
CSJ se ordinauerunt BD. 
om. ips"os BD. 

16 subiecti (uersum om. A) B dr.. 
talibus BD d g r. 
laboranti] + in uobis BDO g. 

17 adimpleuerunt BD. 

18 enim] autem B. 
om. qui B. 
huius modi D. 
om. sunt B. 

19 c\} Aquila multum B dr. 
Priscilla ABDO d g Ambst. 

ea quae in domo eorum est ecclesia B 4- apud quos 
etiam (et g) hospitor BD (a. q. inhospitor O) d g. 

22 Dominum] + nostrum BDO. 

23 Domini] + nostri BDOZ r. 
om. lesu B (hahet Ambst.''*®). 

2 Cor. i 1 CV3 Christi lesu ABDOZ. 

cvj Sanctis omnibus AB ADFOZ d g r Aug. (omnibus 
Sanctis Ambst.*'*^). 
2 om. patre — Christo B. 

4 om. omni B. 
pressura BD Ambst. 

qui alt] + sunt BD Ambst. Ambr. 

angustia BD. 

om. sunt BD Ambst. Ambr. 

5 abundat] + etiam BD Ambst. Ambr. + et g. 

6 angustiam patimur B, (ex angustiamur) Ambst.*'*® 

(pressuram patimur Ambst.). 

Nothing could be clearer than the meaning of this evidence. 
Here we find in the Ambrosiaster interpolation sixty dififerences 
from the Vulgate text. Of these sixty, no less than fifty-four are 


reproduced in the Balliol manuscript. Of these fifty-four instances, 
thirteen are supported by no other known authority. It is clear 
therefore that the complete Pelagian codex from which the inter- 
polation was made exhibited already, in the period between A.D. 400 
and 550, a Biblical text which is almost exactly reproduced in the 
Balliol manuscript. This complete Pelagian codex was almost cer- 
tainly an older MS than that to which the Vatican fragments be- 
longed. There is no reason to doubt that both these early MSS 
were Italian. Yet for the complete text to which they witness we 
are now dependent entirely on an Irish-Italian MS of the middle 
of the fifteenth century. 

The net result of our investigation up to this point, is that 
the text of the Epistles of St Paul commented on by Pelagius is 
represented in its fullness by one manuscript only, the Balliol MS. 
It is not strange that here and there it bears traces of harmoniza- 
tion with the Vulgate, but these are as nothing compared to the 
systematic alteration in the same direction to which the Reichenau 
manuscript bears witness. We shall find on occasion, however, that 
the Reichenau MS helps us to recover the true text employed by 
Pelagius, where the Balliol MS has suffered this deterioration. 

There is another striking fact to which attention must again 
be called at this stage, that is, the relationship of the Book of 
Armagh to the text employed by Pelagius. Of the sixty readings 
referred to above, thirty-one are found in the Book of Armagh (D) 
also, one being peculiar to Ambst.'"'®''""'- and D, namely huius modi 
(1 Cor. xvi 18). No other outside authority is so close. Even the 
Reichenau MS agrees only six times in this section with B, a fact 
which is however somewhat discounted by the omission of whole 
verses of scripture fironi the former manuscript. The real proportion 
of the agreements between A and B is better exhibited by the 
table of readings already published^ where they agree in twenty- 
eight out of forty-four passages, against the Vulgate. Both the 
Reichenau MS and the Book of Armagh, therefore, exhibit Pelagius' 
Biblical text of St Paul's Epistles, after it has suffered harmonization 
with the Vulgate, but the harmonization is in each case independent 
of the other, and it is also partial, not complete. The Pauline text 
in the Book of Armagh is, then, in my view, taken out of a copy of 
1 Brit. Acad. Proc. vol. vii (1915—16) pp. 271—274 (Epistle to the Eomans only). 


Pelagius's commentary on the Epistles, after that text had been 
partially harmonized with the Vulgate. It will not be unsafe, 
therefore, to call in the help of D in doubtful cases. 

In order to show as strikingly as possible the close relationship 
between the text in the Booh of Armagh and that provided by the 
Balliol codex of Pelagius, it Avill be best to select the peculiarities 
of the Armagh text that have been signalised by two investigators, 
to whom our present problem was necessarily quite unknown. 

Samuel Berger in his Histoire de la Vulgate gives a list of in- 
terpolations in the New Testament text of the Book of Armagh (Dy. 
I propose to repeat it here, adding the text of the Balliol MS of 
Pelagius (B) at each point. 

Rom. iiii 18: tamquam stellae caeli et tamquam harena^ quae 
est ad Oram maris non denumerabitur a multitudine. 

This is omitted both by A and B (= vg), which indicate an 
earlier stage. 

Rom. XV 30: ut solicitudinem inpertiamini in orationibus 
uestris ad dominum pro me. 

B reads: ac sollicitudinem inpertiamini mihi in orationibus 
uestris ad dominum. The pro me in D is doubtless taken from the 
Vulgate, which reads ' ut adiuuetis me in orationibus pro me ad 

1 Cor. vii 35: ut sit frequens oratio uestra ad dominum sine 
ulla occupatione. 

All this is absent from A and B alike ; its origin is a matter 
of considerable interest, as it differs from the Vulgate ' et quod 
facultatem praebeat sine impedimento dominum obseruandi^' 

1 Cor. xi 24: quod tradidi pro uobis. 

B has quod pro uobis tradetur (= vg), A has simply pro uobis. 

Gal. ii 2: qui existimabantur esse maiores. 

This is a very instructive case. B reads qui existimantur esse 
maiores. The Vulgate has qui uidebantur simply. 

Eph. ii 5: peccatis et concupiscentiis uiuificauit nos in Christo 

1 p. 32 n. 3. 

- Actually hariena (Gwynn). 

' This celebrated verse is the subject of a discussion by Jerome (ad louin. i 13), 
who blames the Latin MSS which, like our A and B, have no translation of the Greek 
here. See De Bruyne in Revue Biblique nouv. s6r. vol. xii (1915) pp. 364 f. 


cuius gratia siimus salui per fidem facti, et simul suscitauit 
simulque fecit sedere. 

B reads peccatis conuiuificauit nos Christo {om. A) cuius gratia 
estis salui facti per fidem et simul suscitauit simulque fecit 
sedem (sedem is a palaeographical error, due to misreading of the 
insular script). The thick type indicates differences from the Vulgate, 
as elsewhere here. 

Col, iii 8 turpiloquium de ore uestro non procedat. 

B reads turpiloquium. de ore uestro non procedat; identicals 

1 Tim. ii 6 temporibus suis datum est. 
B is here identical with D. 

2 Tim. iiii 5 tu uero sobrius esto, in omnibus labora. ..iam enim 
ego immolor. 

B reads tu uero uigila in omnibus... ego enim iam delibor 
(almost = vg). 

Let us now examine Dr Gwynn's nine instances of 'singular 
readings of D^.' Of these instances three have already come under 
our view above, namely Rom. iiii 18*; 1 Cor. vii 35 and Eph. ii 5. 
There remain, therefore, only six instances to consider: 

Rom. i 27 exercentes. 

B agrees: vg has operantes. 

Rom. xii 9 execrantes. 

B agrees: vg has odientes. 

1 Cor. xi 4 supra caput habens uelamen. 

Here all Pelagian authorities have uelato capite (= vg). It may 
be that we ought to put D's reading in the text of Pelagius. 

2 Cor. vii 11 sinceres. 

B agrees: vg has incontaminatos. 

Eph. vi 18 tolerantia for instantia of vg. 

B omits altogether the clause in omni instantia et obsecratione, 
but both D and B have a semper there which is absent from the 

Phil, i 27 certamen ineuntes cum fide. 

B has here conlaborantes fidei (= vg). 

' Here Berger has gone badly wrong : he fails to give the uestro of D, and at the 
same time he adds cum actihus, which according to the Irish custom really belongs 
to expoUate ueterem hominem. 

2 Liber Ardmachanus, pp. ccxx f. Of the five dittograph readings (p. ccxxi) of 
D, B (or A) has three, of the six mere blunders only one, or rather a half of one. 

3 8 is given by a slip for 18. 


There can be no doubt of the close relationship of the two texts. 
In all, fifteen testing passages have been considered. In five of these 
passages B agrees with the Vulgate against D, and it may be that 
here D has preserved the true Pelagian text. In eight and a half^ 
others the connexion between B and D is the very closest. In the 
remaining one and a half passages B and D differ from one another 
and from the Vulgate. In 1 Cor. vii 35 B clearl}^ indicates a more 
primitive text than D, as it gives the true Old-Latin reading, and 
at Eph. vi 18 the omission of five words from B may be accidental. 

External Confirmation of the use of the D type of text hy Pelagius. 
Epistula ad Demetriadem 

Having asked myself the question whether other works of 
Pelagius would shed any light on the type of Pauline text he used, 
I turned to the Epistula ad Demetriadem. This was at a time when 
I was fortunately in touch with a good manuscript of this work, 
namely Codex Augiensis cv (saec. IX in.) at Karlsruhe-. My friend 
Mr E. J. Turner* and I collated the MS as far as the middle of c. 26. 
In it the letter is attributed, as sometimes also elsewhere, to Julian, 
and it commences on fol. 15 a. It may be remarked incidentally 
that a critical edition of this letter will differ considerably from the 
published form, as like other heretical works, it has been neglected. 
Here I will call attention merely to one lengthy quotation from the 
Epistle to the Romans in c. 4, which reads thus according to the 
Reichenau manuscript: 

Rom. ii 14 — 15 cum enim gentes quae legem 

non habent 
naturaliter quae legis sunt 

huius-modi legem non 

habentes ipsi sibi sunt lex 
qui ostendunt opus legis 

scriptum in cordibus suis 

1 The 'half is Eph. vi 18 where B and D agree in semper, but differ otherwise. 

- Now described by Holder, Die Reichenauer Handschriften i Bd. (Leipzig, 1906) 
pp. 267 ff. 

3 Formerly Major Scholar of Trinity College, Cambridge, and long of H.M. India 
Office. The collation was made 16 Aug. 1906, 


testimonium reddente eis 

conscientia eorum 
et inter se inuicem accusan- 

tibus cogitationibus aut etiam 

What are the authorities for the non- Vulgate readings here? As 
given by Wordsworth and White they are: 

huius T dem t Orig. |. 

eis Orig. ter. 

eorum c Sedul.' 

' accusantibus cogitationibus ' appears to be unique, but it 
is close to 'cogitationibus accusantibus' D g^ Orig. f 
(Ambst.) Arabr. Hier. Aug. bis Sedul.^ 

defendentibus D Orig. f gl Ambst. Ambr. Hier. Aug. Sedul.^ 
The evidence here speaks for itself, I think. I have not pursued 
the matter further, but it will be quite instructive to compare the 
many quotations in the Letter to Demetrius, when once it is critically 
edited, with the Pelagian text as published in my second volume, 

De Induratione Cordis Pharaonis 

A tractate which circulated under this title in the Carlovingian 
period as a work of Jerome, has been recently rediscovered by Dom 
Morin in six manuscripts, of which four at least are British in origin. 
Dom Morin's edition will be published in the second volume of his 
Etudes, Textes, Decouvertes. Meantime, we have to depend on a 
considerable account of the work which he has published in the 
Revue Benedictine^. He considers it probable that the author of 
the tractate lived towards the beginning of the fifth century, as 
the Pelagianism it displays is in a very crude state. It will be of 
some interest to examine the character of its quotations from 
the Epistle to the Romans. Whether it be by Pelagius or not, 
it is interesting to observe that the word induratio is found 
also in the Pelagius commentary at 1 Tim. ii 4 — de induratione 

If the suggestion made below, p. 148, is right, Sedul. = Pelag. 
2 Vol. XXVI (1909) pp. 163 — 188 : see also his Etudes, Textes, Decouvertes t. i 
(Maredsous and Paris, 1913) p. 24. 

'^ See Benoist-Goelzer for other exx. (from Paucker). 


Rom. vi 16 oboedistis (oboeditis vg) vg'=°*^'^- p'^""- D c dem gr^t 
Orig. Ambst. Ambr. B. 
viii 30 magnificauit (glorificauit vg) vg*'°'^*'- ™'^''- D c d* dem 
Ambst. A (honorificauit B). 
32 qui unico filio suo non pepercit (qui etiam f. s. n. p. vg 
A) Orig. semel Aug. semel (qui filio s. proprio 
n. p. B). 
viiii 11 bonum uel malum (bonum aut malum vg A) D Aug. 
semel B. 
18 cui (cuius vg A) y^°^^- n^"'*- D c d Ambr. Orig. ter 
Ambst. Aug. semel Beatus B. 
obdurat (indurat vg AB) DF'^sR Ambr. Aug. 
{sempei') Beatus Sedul. 
20 cur me ita fecisti (quid me fecisti sic vg AB epist. 
ad Demetr.) unique. 
This investigation does not seem to prove anything, but it can 
hardly be said to contradict the possibility that the De Induratione 
Cordis Pharaonis is the work of Pelagius himself. 

Relation of the Pauline text used by Pelagius to the quotations 
in Gildas (a.d. 500— 570)i. 

Rom. i 21 f. cognouerunt with A. 

magnificauerunt with DB. 
obcaecatum with D. 
om. enim. 
25 mendacium with DA Ps.-Hier. 

07n. amen with AB and one family of Ps.-Hier. MSS. 

28 notitia tvith A. 

29 inpudicitia fornicatione. 

30 inoboedientes. 

31 insensatos with B. 

om. absque foedere with DAB. 
sine misericordia sine adfectione. 

I have used the edition of Mommsen in Chronica Minora vol. in (M.G.H. auct. 
ant. t. xiii) (Berol. 1898), not yet accessible to Haddan and Stubbs, Councils and 
Ecclesiastical Documents, etc. vol. i (Oxford, 1869) pp. 170 — 185. There is a reference 
to Gildas's Biblical quotations in Zeitschr. f. celt. Philol. Bd. iv pp. 572 f . by the late 
Dr Hugh Williams. 


Rom. i 32 non solum] + qui with DAB. 
faciunt ea with DB. 
etiam with Ps.-Hier. as printed. 
+ qui with DAB. 
ii 5 tu autem secundum duritiam tuam with D. 
cor inpaenitens with DB. 
6 sua with DB Ps.-Hier. cod. 

11 acceptio personarum. 

12 om. et {pr.) with DB and one family of Ps.-Hier. 

om. et alt. 
vi 2 adhuc] iterum with DB, 
viiii 3 om. ipse ego mfA D. 

xi 3 add. et a^iie altaria with Aug. 
xiii 12 adpropinquauit with D (def B) and one family of 
Ps.-Hier. MSS. 
induamus with D (def B) and one Ps.-Hier. MS. 
14 induite with AB. 

concupiscentiis with DB. 
1 Cor. iii 10 ff. alter with B. 

ow. autem with B. 

om. positum est qui. 

lesus Christus with D. 


super with D^ 

om. fundamentum. 

aurum] + et with B. 

unum quodque {pr.). 

enim] -I- domini with DB (def A). 

declarabit] -f- illud. 

qui superaedificauerit (quod superaedificauerit 

om. ipse autem... per ignem. 
(18) apud. 

V 7 expurgate] + igitur with D and one family of 
10 exrre with DB and one family of Ps.-Hier. 
1 I take it that Dr Gwynn has here wrongly resolved the insular symbol. 


1 Cor. V 1 1 is qui] quis with DB and one MS. of Ps.-Hier. 
nominatur frater. 

] + et with D. 
huiusmodi with D. 

cibum] + quidem with DB and one family of 
Ps.-Hier. MSS. 
2 Cor. iiii 1 f. administrationem with B and one family of Ps.- 
Hier. MSS. 
deficiamus with D and Ps.-Hier. MSS. 
abiciamus ivith some Ps.-Hier. MSS. 
xi 13 ff. pseudo-apostoli] -I- sunt luith D and one family of 
Ps.-Hier. MSS. 
magnum igitur. 
uelut ministri] ut angeli. 
Eph. iiii 18 f uia ivith D. 

eorum Ps.-Hier. codd. ple)\ 
omnis inmunditiae. 

in auaritia] et auaritiae with DAB Ps.-Hier. cod. 
V 17 f dei with D Ps.-Hier. cod. 

replemini with one family of Ps.-Hier. MSS. 
spiritu] + sancto with D and two Ps.-Hier. MSS. 
Phil, i 8 cupimus unum quemque uestrum in uisceribus 
Christi esse (c/. cupio with one family of Ps.- 
Hier. MSS). 
1 Thess. ii 5 ff. fuimus apud uos aliquando. 

om. deus testis est iuith one family of Ps.-Hier. 

gloriari with one family of Ps.-Hier. MSS. 
possumus luith some Ps.-Hier. MSS. 
honori with D and one family of Ps.-Hier. MSS. 
ceteri apostoli Christi with one family of Ps.-Hier. 

sicut paruuli (with one family of Ps.-Hier, MSS). 
tamquara] uel tamquam. 
paruulos with one family of Ps.-Hier. MSS. 
uobis tradere with one Ps.-Hier. MS. 
om. dei. 


1 Thess. iiii 2 ff. dederim with most Ps.-Hier. MSS. 

om. per dominum lesum with A and one family 

of Ps.-Hier. MSS. 
om. enim. 
ut sciat] et sciat. 
uas suum. 

honore et sanctificatione. 
(6) ut] et with one family of Ps.-Hier. MSS. 

om. siciit...sumus with A and one family of Ps.- 
Hier. MSS. 
inmunditiam with DAB and one family of Ps.- 
Hier. MSS. 
sanctificationem with DAB. 
Col. iii 5 f libidinem] + et. 

om. et auaritiam...seruitus (but not on p. 66). 
super] in. 
1 Tim. i 15 (iiii 9) sermo] -1- est. 

iii 1 fif. desiderat 1°] cupit. 

episcopum] huiusmodi ivith DB. 
~hospitalem ornatum 
om. doctorem. 

domum suam bene regentem with (s. d.) D. 
uino multum. 
(10) om. Et. 
vi 3 ff. sermonibus sanis. 

superbus] + est with Ps.-Hier. as printed, 
languescens with B. 
erga with one family of Ps.-Hier. MSS. 
(5) om. et luith B and one family of Ps.-Hier. MSS. 
17 ff. noli superbe s. uel (c/. D and B). 
nobis] tibi. 
omnia] multa. 
om. ad fruendum. 
adprehendant] habeas. 
2 Tim. ii 4 f placeat ei. 
s. p. 10 


2 Tim. iii 1 ff.^ enim scitote. 

periculosa] pessima |. 
et erunt] erunt enim. 

om. sine pace criminatores (om. sine pace B). 
uoluptatum | loith B and some Ps.-Hier. MSS. 
quidem speciem tvith B and one family of Ps.- 
Hier. MSS. 
(8) autem] enim ^uith one family of Ps.-Hier. MSS. 
hi] isti with B and nearly all Ps.-Hier. MSS. 
Tit. ii 8 sanum] -f- habens. 

nullum malum habens ivith one family of Ps.- 
Hier. MSS. 

These statistics are very significant, especially as Gildas's quota- 
tions are long, and therefore copied fi-om a Biblical MS. There are 
out of 121 variants, 39 agreements with D, 13 with A, 33 with B, 
and 43 with one or more MSS of Pseudo- Jerome. It is quite evident 
that Gildas, about four generations after Pelagius, employed a text 
substantially identical with his and that of the Book of Armagh. 
Is it too much to argue that in Pelagius we find the textual form 
of the Epistles used in Britain in the pre-Vulgate period, and for 
long afterwards, that here we have the oldest form in which the 
Pauline Epistles were read in the British Isles, the missing British 
form of the Old-Latin ? 

A presumption having been established that the Pauline text 
employed by the Briton or Irishman Pelagius, by the Briton 
Gildas, and by the scribe of the Book of Armagh, is really a British 
text with definite characteristics of its own, it becomes a matter of 
no little interest to enquire whence this text came. The two re- 
cognised routes of travel from the Continent to Britain were by 
Boulogne and Richborough on the one hand, and from the Rhine 
to Colchester on the other-. It cannot be denied that there was also 

1 I purposely omit the quotation of these verses in ihe fragm. epist. Gildae pub- 
lished by Mommsen on p. 86 of his edition. Whether the fragment be by Gildas or 
not, it appears that the quotation is not taken from a copy of the Epistles, but from 
Jerome on Zephaniah, chap, ii (ed. Vallarsi, t. vi 713 A of the 4to edition). 

2 F. J. Haverfield in Cambridge Medieval History vol. i p. 370. 


direct intercourse between Spain and Gaul on the one hand, and 
Cornwall, Wales and Ireland on the other, but the other routes 
were the most important. We have then to ask what continental 
texts are most likely to have been transmitted to Britain. If we try 
such Spanish texts as are accessible to us, those of m, t, 'Priscillian' 
and Gregory of Elvira, we shall not I think discover many notable 
coincidences between the Pelagian text and these. A comparison 
with the quotations in Hilary of Poitiers will be hardly less dis- 
appointing'. But when we come to the quotations in Ambrose, the 
situation is changed. It is a delicate question whether Ambrose 
used at Milan the type of scripture text in use in his native church 
at Treves; but as he was brought up in a Christian family, it seems 
to me not improbable that this very ardent student of scripture 
used the text current at Treves. It is indeed possible that there 
was no great difference between the Treves text and the Milanese 
text. The resemblances between the Ambrose and Pelagius texts 
are set forth below. I have been careful to select, for the purpose of 
forming a judgment, only lengthy quotations from Ambrose in a 
modern critical text. I have gone through the six volumes of 
Ambrose already published in the Vienna series, and gathered 
together such passages as seem to have been copied straight from 
a Pauline codex, and then brought these quotations face to face 
with the Pelagian text, here denominated B. It will be seen that 
there are considerable differences between the two texts : I do not 
seek to argue for their identity. But what I think may be regarded 
as fairly established is this, that there is a real kinship between 
these two West-European Pauline texts such as one will seek in 
vain elsewhere. Each has its distinctive characteristics, but they 
spring from the same root. Is it too much, then, to hazard a con- 
jecture that Britain got its Pauline text from the Rhine country 
via its oldest Roman colonia, Camulodunum (Colchester)? 

An examination of the valuable lists in the introductory part 
of Dr Gwynn's Liber Ardmachanus shows a large number of passages 

^ Hilary, as a consummate Greek scholar, seems often to have translated 
direct from the Greek. Also, we lack a critical edition of most of his works with 
an index suited to modern requirements. So far as Pauline references in his com- 
mentary on the Psalms are concerned, I have endeavoured, in J.T.S. vol. xviii 
(1916—17) pp. 73—77, to supply the want of an index in the Vienna edition. 





where D has the support of Ambr., and if there be many where it 
has the support also of Sedulius Scottus, the reason for this is no 
doubt the adoption of Pelagius's Biblical text by Sedulius along with 
a large proportion of the Pelagian comments. In fact my investiga- 
tions have led me to believe that anything in Sedulius's Biblical text 
that differs from the Vulgate is Pelagian, and that one might well 
substitute Pelag. for Sedul. in already published works that quote 
Pauline Latin codices. These works convey the impression that 
Sedulius in the ninth century used an Old-Latin text of his own. 
There is in fact in his Collectaneum, as I have proved, hardly one 
column of original matter^ ! 

Ambroses disagreements with the Vulgate, paralleled 
by Pelagius's text 

Vienna edition, 

Rom. i 15 


est promptus sum B 

promtum est vg 

fug. saec. pp. 173, 174. 

Rom. ii 14 


cum vg 

parad. p. 296. 

Rom. V 4 

autem B 

uero vg 

ps. 118 p. 205. 

Rom. vi 4-7 

in morte 

in mortem vg 

lacob p. 52. 

quern ad modum B 

quomodo vg 

parad. p. 303. 

ille surrexit (Chr. s. B) 

surrexit Christus 



mortuis vg 


gloriam vg 

propriam suscitatus 

patris vg 

eius gratia 

in nouitate uitae 



ambulemus vg 

facti om. B 

facti vg 

hoc om. 

hoc vg 

confixus est cruci B 

crucifixus est vg 

Rom. vi 16 

in mortem B 

om. vg 

lacob p. 1 1 , 

Rom, viii 38 f. 

confido B 

certus sum vg 

ps. 118 p. 172. 
ps. 43 p. 288. 

potestates | 

principatus vg 

praesentia f 

instantia eg 

tseparare nos poterunt 
\ (possint) 

poterit nos separare vg 

domino nostro om. \ 

domino nostro vg 

1 See J.T.S. vol. xviii (1916—17) p. 228, I regret that I did not soon enough 
realise the significance of the lemmata in Sedulius Scottus, having my mind too 
exclusively concentrated on his commentary. 





Rom. xi 1 f. 

heredltatem suam # 

Rom. xi 25 f. 

Israhel contiglt B 
Rom. xiii 4 

egit {def. B) 
Rom. xiii 8 f. 

suum {def. B) 

scriptum est enim {def. B) 
1 Cor. i 4 f. 

dei nostrl 

Ipso B 
1 Cor. iii 18f. 

sap. esse inter uos # B 

1 Cor. iiii 11—13 


1 Cor. vii 29 ff. 


jita slnt ac si nonl 

I habeant | 

hoc mundo utuntiir 
1 Cor. xii 23 

esse membra {def. B) 

I honestatem abundanti- 

\ orem {def. B) 
1 Cor. XV 22 ff. 

sicut enlm B 

in aduentum (-u B)| 
eius credlderunt B 


et uirtutem om. 

1 Cor. XV 51 f. 
nobis om. B 

Jnon omnes autem (non 

( omnes B) 

jin atomo in momento 

I oculi 

canet enim om. B 

2 Cor. i 3 ff. 

nos exhortatur 
nostra om. 

(sunt in omni tribnlati- 
( one (angustia B) B 
et ipsi aduocamur 
(5) et om. 

abundat etiam B 

(qui ii 
] eiu 

populum suum vg 

contigit in Israhel vg 

a git vg 

om. vg 
nam vg 

dei vg 
illo vg 

inter uos sap. esse vg 

in vg 

caedimur vg 
purgamenta vg 
peripsima vg 

ut vg 

jtamquam non haben-| 
I tes sint vg ( 

utuntur hoc mmido vg 

membra esse vg 
Jabundantiorem hones- 
I tatem vg 

et sicut vg 

in aduentu eius vg 

■ cum (2°) vg 
hahet vg 

habet vg 

sed non omnes vg 

jin momento in ictu 
I oculi vg 
hahet vg 

totius vg 
consolatur nos vg 
tribulatione vg 
hah. vg 
Jin omni pressura sunt 

I ^5' 

exhortamur et ipsi vg 

habet vg 

abundat vg 

Vienna edition, 


ps. 118 p. 34. 
ps. 43 p. 281. 

ps. 118 p. 424. 

Cain p. 391. 

ps. 118 p. 498. 

ps. 118 p. 39. 

ps. 118 p. 493. 
ps. 36 p. 75. 
ps. 43 p. 263. 

ps. 118 p. 153. 

fug. saec. p. 199. 

de Noe p. 428. 
ps. 118 p. 278. 

exam. p. 121. 

ps. 118 p. 225. 




Vienna edition, 


2 Cor. vi 11 ff. 

coartamini B 

angustiamini (2°) 

ps. 118 pp. 124, 80, 
154 etc. 

2 Cor. vi 14 

et ante noUte B 

om. vg 

portio 1 

partici patio vg 

2 Cor. vi 16 


illis vg 

deambTiIabo | 

inambulabo vg 

+ inipsis | (Ms \) 

om. vg 

2Cor. xiiSfF. 


huius vg 

parad. p. 265 
ps. 36 p. 102. 

uerba inefifabllia f 

arcana uerba vg 

(5) huius 

eius vg 


nihil vg 

(6) si enim 

nam etsi vg 

stultus B 

insipiens vg 

nam ueritatem dlco (B) 

Jueritatem enim dicam 

1 ^'9 

Gal. iii 16 

repromissiones ^ 

promissiones vg 

parad. p. 278 
ps. 36 p. 118 

tamquam | B 

quasi vg 

sicut 1 B 

quasi (2°) vg 

Gal. iiii 21 ff. 

sub lege uolentes esse B 

(qui sub lege uultis 
1 esse vg 

Abr. p. 523.. 

quod B 

quouiam vg 


habuit vg 

Jsed is quidem qui (sed is 
1 quiB) 

sed qui vg 


repromissionem vg 

□am baec 

haec enim vg 

quod est Agar 

quae est Agar vg 

Gal. V 22 

patientia B + 

longanimitas {vg) B 

parad. p. 323 
cf. ps. 36 p. 78. 

bonitas om. \ (om. benig- 

habet vg 

nitas B) 

fides om. ^ 

habet vg 

dilectio {in fine) ^B 

om. vg 

Eph. ii 14 ff. 

inimlcitias B 

inimicitiam vg 

Abr. p. 585, 


decretis vg 


condat vg 


ipsum vg 

uno nouo homine 

junum nouum hominem 
1 ^,9 

Jut reconciliaret Ps.-Hier. 

1 codd. 

et reconciliet vg 


ambos vg 

iuimicltias B 

inimicitiam vg 




Eph. vi 12 

Trectores huius mundi 
J (mundi huius rectores B 
t Ambr. ps. 1) 


nequitiam spiritalium ^ 

quae sunt it 
Phil, i 23 f. 

dissolui enim cupio 

Phil, ii 6 ff. 

esset in forma del (etc.) 


exaltauit ilium deus | 


eius ^ 

Christus 07)i. B 
Phil, iii 7 f. 

lucra fuerunt B 

duxi B 

Christi lesu B 


passus sum 

Col. i 13ff. 

claritatis {Schenkl, etc. : 

? errore pro caritatis B) 

redemptionem et 
inuisibilis et ^ 
ipse J 

uniuersae ^ 
in quo 
sine uisibilia 
sedes B 
Col. ii 18f. 

uidet Ps.-Hier. codd. 
extollens se B 
mente B^ 

colligationes B 

principes vg 

aduersus (S") vg 
mundi rectores vg 

contra vg 

spiritalia nequitiae vg 

om. vg 

desiderium habens dis- 
solui vg 

magis vg 

in forma dei esset vg 

habitu vg 

deus ilium exaltauit vg 

donauit vg 

lesu vg 

habet vg 

fuerunt lucra vg 
arbitratus sum vg 
lesu Christi vg 
omnia vg 
feci vg 
arbitror ut vg 

dilectionis vg 

redemptionem vg 
inuisibilis vg 
qui vg 
omnis vg 
quia in ipso vg 
condita vg 
uniuersa vg 
caelis vg 
uisibilia vg 
et vg 
throni vg 

uidit vg 
ambulans vg 
sensu vg 
totum vg 
nexus vg 
coniunctiones vg 

Vienna edition, - 


parad. p. 313 

ps. 1 p. 33. 

cf. ps. 36 p. 113. 

ps. 118 p. 501 
cf. ps. 47 p. 354 
ps. 61 p. 390. 

ps. 118 pp. 329, 414, 

311, 454 
ps. 35 p. 53. 
ps. 61 p. 382, etc. 

ps. 118 p. 154. 

exam. p. 232 

ps. 118 pp. 52, 212 

cf. ps. 36 p. 100. 

ps. 118 p. 446. 

1 Cf. in this chapter, p. 154. 




Vienna edition, 


Col. ii 18 f. {cont.) 

copiUatum B 

constructum vg 


augmentum vg 

Col. ii 21 ff. 


tetigeritis B 

tetigeris vg 

Noe p. 480. 

jne adtaminaueritis | (con- 
jne gustaueritis | B 

B)j J jneque gustaueris 1 . ps. 118 p. 323 
f (neque contrectaueris | " ps. 1 p. 23. 

omnia om. \ 

hahet vg 

ad corrupteiam § 

in interitu vg 

(obseruatione religionis B 

superstitioue vg 

) {om. religionis) 

humilitate cordis B 

humilitate vg 

(non in indulgentia cor- 

iet ad non parcendi 
j corpori vg 


1 pons 

Isaturitatem et dillgen- 

saturitatem vg 

1 tiam B 

Col. iii 9 ff. 

uos om. f, Ps.-Hier. codd. 

hahet vg 

ps. 118 pp. 53, 213. 

Induite | B 

et indueutes vg 

(et indutmus |) 

eum om.. f B 

hahet vg 

agnitione'l B 

agnitionem vg 

2 Thess. ii 3 f. 


peccati vg 

ps. 118 p. 60. 

] + et 

om. vg 


perditionis vg 

1 Tim. vi 3ff. 

languescens B 

languens vg 

ps. 118 p. 132. 


blasphemiae vg 


et qui vg 

a ueritate alienorum 

ueritate priuati sunt 


Tit. iii 3 ff. 

et nos om. f 

hahet vg 

ps. 118 p. 489 
ps. 40 p. 245. 

errantes in ^ B 

errantes seruieutes 


dei om. 

hahet vg 

Philem. 12 ff. 

ut B 

id est vg 

ps. 118 p. 314. 

sed sine B 

sine vg 

autem om. B 

hahet vg 

uti om. 

hahet vg 


I should just add, in a word, that I have gone through the 
quotations adduced in my Study of Amhrosiaster, pp. 240 — 257, 
and have found that, side by side with certain coincidences, there 
are many signs of individual character in the Pelagius text. 

1 ' Die Wortf olge des Ambrosius, dass gustare erst an dritter Stella, nach attami- 
nare (oder contaminare, oder contrectare) steht, findet sich meines Wissens bei keinem 
unserer Zeugen, der Plural nur bei wenigen lateiniscben.' (H. Eonsch, Zeitschr. f. 
hist. Theol. 1870, i p. 108.) 


The Pelagian text being then an Old-Latin text, which when 
used by Pelagius was free from all contamination with the Vulgate, 
it is my duty to give some account of its Latinity. 

There is a tendency to give the Divine name in full, where more 
correct texts give it more briefly, e.g. 1 Cor. v 4b domini nostri 
lesu Ghristi for domini lesii; 1 Cor, v 5 domini nostri lesu Christi 
for domini lesu. 

dilectissimus rather than dilectus or carissimus: Eph. vi 21; 
Phil, ii 12; 2 Thess. ii 13; Col. iiii 7. The use of the superlative, 
which has not necessarily got more value than the positive, is a 
feature of certain Old-Latin texts, where it is really a translation of 
the Greek positive. Cf. Burkitt, Rules of Tyconius p. xcii (optimus), 
p. xciv (pessimus), p. xcvi (proodmus). 

dilectissimus is early, being found in k of the Gospels, and 
Cyprian himself affects it^ It is also a feature of the African 
psalter^, whereas the non- African have dilectus. The Latin Irenaeus 
has the superlative twice, the positive once. 

facies (rather than persona) as a translation of irpoa-wirov: 
2 Cor. ii 10; Gal. ii 6; 1 Thess. ii 17. 

hie mundus (rather than mundus simply). This point forms a 
good criterion in the study of Latin Biblical texts. Rom. v 13; 

1 Cor. i 21, 28; ii 12; iiii 9; vi 2 his; vii33, 34; 2 Cor. i 12; Eph. 
ii 12; vi 12; Col. ii 8, 20 bis. 

The addition of hie is on the whole un-African and un- 
Cyprianic^: but see Burkitt, Rules of Tyconius p. Ixxxii. 

inreprehensibilis (rather than sine querella, crimine, etc.): 

2 Cor. vi 3; Eph. i 4; Phil, ii 15; Tit. i 7. 

ipse as a rendering of ovra or avro'i, preferred to ille, is: 
1 Cor. i. 5 ; vii. 20; viii. 12; x. 10; (2 Cor. ii. 17); 2 Cor. viii 2; 
24 etc. 

There is some reason to regard ipse as late African^ Victorinus 
Afer's text of Ephesians shows a partiality for it. It abounds in 
the Latin Irenaeus. 

Watson, Style and Language of St Cyprian pp. 272 f. 
- Capelle, Le Texte du Psautier Latin en Afrique pp. 100, 128. 
^ Hans von Soden, Das Lateinische Neue Testament in Afrika zur Zeit Cyprians, 
pp. 40 etc. 

■* Von Soden, op. cit. pp. 156, 302. 


magnifico (rather than clarifico, glorifico, honorifico, honoro): 
Rom. i 21; viii 30; 2 Cor. viiii 13; Gal. i 24. 

clarifico is definitely African \ while gloynfico, honorifico and 
magnifico are rather European. The Latin Irenaeus shows all, 
except honorifico, in about equal proportion. Magnifico is found 
twice in Tyconius (see the evidence in Burkitt, Rules p. Ixxxix). 

mens (rather than sensusY: Rom. xii 2; Phil. iiii. 7; 2 Thess. 
ii2; Col. ii 18. 

nam (rather than enim): Rom. vi 2; vii 18; viiii 11; xiiii 3, 
10, 15; 1 Cor. iii 3; (x 17; xvi 1); xvi 10; 2 Cor. iiii 17, (18); Eph. 
V 9; Phil, iii 18 etc. 

A predilection for nam appears to be late African rather than 
early African I Victorinus Afer often has nam where the Vulgate 
has enim. In the Latin Irenaeus enim is used almost to the ex- 
clusion of mim. 

pressura (rather than trihulatio): 2 Cor. i 4(1°), 8; iiii 8, 17; 
Eph. iii 13; 1 Thess. i 6; iii 7; 2 Thess. i 4, 6. 

Pressura is much more African than European^; the reverse 
is true oi trihulatio. Dr Gwynn {Liber Ardmachanus p. ccxxi) notes 
that pressura is habitually preferred to tnbidatio by D. 

quasi (rather than tamquam, quemadmodum, uelut, sicut, qiio- 
modo, ut): 1 Cor. iii 1; iiii 9; 2 Cor. vi 10. 

Quasi is the original equivalent in the African Bible ^ but 
quemadmodum and sicut are also founds Victorinus Afer favours 
quemadmodum, as does the Latin Irenaeus, but the latter has also 
some examples of quasi, sicut; the others occur there hardly at all. 
Tyconius has quasi six times, tamquam four times, uelut seven times, 
sicut ten times, quomodo once (twice), ut twice (Burkitt, Rules 
pp. xcvi ff.). 

quia (rather than quoniam, in reported statements, etc.): Rom. 
vii 21; 1 Cor. vi 19; 2 Cor. i 7; vii 8; xi 18; Eph. ii 11; Phil, iiii 
10; 1 Thess. ii 8; 2 Thess. iii 7. 

1 Of. von Soden, p. 239 n., etc.; Capelle, p. 30, etc. 

2 See W.-W. on Rom. i. 28. 

3 Von Soden, p. 233. 

4 Sanday, O.-L.B.T. n p. cxxi; von Soden, pp. 137, 142, 181, etc.; J.T.S. xiv 
(1912 — 18) p. 340; Watson, p. 289; Burkitt, Rules of Tyconius p. xcv. 

» Von Soden, pp. 35, 82, 344. 
* Von Soden, pp. 154 f. 


This is as un-African as possible, quoniam being by far the 
most frequent in the early African Bible \ The latter is almost 
universal in the Latin Irenaeus and in Victorinus Afer. The same 
may almost be said of Tyconius, who uses quia about twenty times> 
but not nearly so often as quoniam (Burkitt, Rules p. xcvi). 

uerbum (rather than sermo) is definitely European, and late 
African: Phil, iiii 17; 1 Thess. i 5, 8; ii 5; 2 Thess. ii 2, 15, 17; 
iii 1 ; Col. iiii 3. 

Sermo is definitely African-. It is much more frequent in the 
Latin Irenaeus than uerbum. Sermo and uerbum occur with almost 
equal frequency in Tyconius (Burkitt, op. cit. pp. c, civ). 



If the preceding argument is valid, Pelagius made no use of the Vulgate 
of the Epistles of St Paul, and it cannot be proved that he had any knowledge 
of it. As, however, I was responsible for the statements, made at the end of 
1906^, that 'the text [in the Reichenau MS] on which the commentary is based 
is the Vulgate.. ..If I am right about this Karlsruhe [Reichenau] MS. ..we are 
provided by it with an authority of the highest value for constituting the 
Vulgate text of Paul's Epistles. Perhaps it would not be wrong to say that 
we are thereby provided with the very highest authority which exists...,' it is 
right that I should explain how my view in regard to this matter came to 

In the course of collating the Reichenau MS with the published Pseudo- 
Jerome, which is substantially Sixto-Clcmentine Vulgate, I was constantly 
impressed by the fact that the Reichenau MS supported either the Amiatinus 
or the Fuldensis of the Vulgate or both together, against the Sixto-Clementine 
text. The phenomenon recurred so frequently that at the middle of First 
Corinthians I ceased to note it. It was obvious that I was in presence of a 
very early form of the Vulgate text. There is no need to retract this opinion. 
Except in certain passages which approximate rather to D {Book of Armagh), 
the Reichenau MS of Pelagius provides an almost complete Vulgate text of the 

1 Von Soden, pp. 81, 154. 

2 Von Soden, pp. 71 f., 238; Capelle, pp. 31, 142. 

^ These remarks are quoted from the first of my two papers before the British 
Academy, delivered on Dec. 12, 1906, and printed in their Proceedings vol. ii 
pp. 425 f. 


Epistles in a very early form. Yet it was clear even in those days that Pelagius's 
text was not Vulgate throughout. I should now explain the Reichenau form 
of text as due to the almost complete substitution of a Vulgate text for the 
older type of text employed by Pelagius himself. But from internal evidence 
it is clear that the Reichenau MS must have been copied direct from a manu- 
script of the fifth or sixth century i. As early therefore as the fifth or sixth 
century the Vulgate must have been substituted almost throughout one MS 
of Pelagius's commentary for the text used by Pelagius himself. These substi- 
tutions will remain, therefore, an authority for the text of the Vulgate, and 
one, too, of the very highest value, at least as old as, and possibly a good deal 
older than the Codex Fuldensis (a.d. 546). 

When I expressed my original opinion in 1906, my list of Pelagian manu- 
scripts was almost as full as it is now. Naturally I deferred my examination 
of the latest MSS till I had already studied the earlier. It never occm-red to 
me then that the character of two fifteenth century MSS whose existence in 
Oxford I had noted in 1905, could overturn my original opinion. As these MSS 
bore the name of Jerome, and the catalogues had not observed that the text 
was considerably shorter than the published form, as well as seriously diff'erent 
from it in other respects, it was with the greatest surprise that in July 1913, 
on borrowing the MSS themselves, I discovered that they were of the same 
brevity as the Reichenau MS and offered an older type of Biblical text than it, 
in spite of the fact that they were vitiated by the name of Jerome in the title^, 
A complete collation of the Balliol MS made in 1914 revealed the character of 
the Biblical text used by Pelagius, and I tried to put myself right with the 
public at once by publishing a little note in the Journal of Theological Studies 
for October 1914 called 'Pelagius and the Pauline Text in the Book of 
Armagh 3.' 

The mischief was, however, done. Scholars had been kind enough to support 
the view of my first paper, and my later discovery was made just too late to 
become available to Dr Gwynn for his Liber Ardmachanus. Yet I cannot alto- 
gether regret the mistake, because it has at least helped to give the impulse to 
a good deal of valuable publication by French and Belgian scholars, and it is 
the chief object of this note to call the attention of British students to their 

There reached me in August 1914 a privately printed article by Dom 
Donatien de Bruyne, O.S.B. (who will always be remembered for his identifica- 
tion of the Latin Marcionite prologues to the Epistles of St PauH), under the 
title: 'Etude sur les Origines de Notre Texte Latin de Saint Paul.' This 

* See below for the proof, pp. 206 ff. 

- The Merton MS 26 was examined by Dom Morin as long ago as 1888 (see 
Anecdota Maredsolana iii 3 [1903] p. 194), and 1 gathered from him in 1913 that 
he had had some suspicion as to the facts. 

^ Vol. XVI p. 105. The Great War had already broken out. 

■* Revue Binidictine xxiv (1907) pp. 1 ff. 


article was afterwards published, with slight modifications, in the Revue Biblique 
for October 1915. In it Dom de Bruyne displays all the acute power of reason- 
ing we have learned to expect from him. He reminds us how uncertain is the 
question of the Vulgate of the New Testament ajjart from the Gospels. We 
have Jerome's triple statement that he revised the whole New Testament, but 
we have no Hieronymian New Testament prologues except for the Gospels, and 
Jerome himself never, it seems, makes any use of the Vulgate of the Pauline 
Epistles. As Pelagius is the earliest author to use the Vulgate of the Ej^istles \ 
and as the general Pelagian prologue appears in practically all Vulgate copies 
of the Epistles, sometimes under the name of Pelagius himself ; as, further, a 
preface is the mark of an edition, the author of the Vulgate of the Epistles is 
none other than Pelagius himself ! 

It is not necessaiy to follow his argument further. It is based in part on 
a necessarily imperfect knowledge of the manuscripts containing the Pelagius 
commentary in one form or another, and is, in my opinion, as regards the three 
editions of the Vulgate, really fallacious. I feel, however, that I owe Dom de 
Bruyne and others an apology for the measure of responsibility that belongs 
to me. 

Naturally, a sensational view like that of Dom de Bruyne was not allowed 
to pass unchallenged even in the midst of the Great War. The Abbe Eug. 
Mangenot, professor of the Institut Catholique at Paris, published in the Revue 
du Clerge' fran^ais in 1916 an article entitled 'Saint Jerome ou Pelage editeur 
des Epitres de Saint Paul dans la Vulgate 2.' His article is an able and learned 
exposition of what might be called the traditional view, that Jerome revised 
the whole New Testaments Father M.-J. Lagrange of Jerusalem, author of 
two valuable commentaries, one on the Epistle to the Romans and the other 
on the Epistle to the Galatians (1918), has in recent years made a profound 
study of the Vulgate text of the Epistles of St Paul. Particular attention must 
be called to the article entitled : ' La Vulgate latine de I'Epttre aux Galates et 
le texte grec' published in the Revue Biblique for 1917'*, which followed a 
corresponding article on the Epistle to the Romans in the same review for 1916S 
Alive to the difficulties which Dom de Bruyne has posed afresh, he seeks to 
solve them by the view that Jerome is indeed the author of the Vulgate of the 
Epistle to the Galatians, but that its date falls between the date of Jerome's 
commentary on Galatians (384) and 392 (the date of the De Viris Inlustrihus). 
Lagrange also shows from De Bruyne's own data that Pelagius employed an 
Old-Latin text of Galatians^, and absolutely denies Pelagian authorship of the 

^ Here of course he depends on my 1906 argument. 

^ Revue du Clergi frangais, 1^" Avril et l""" Mai 1916 (Paris, Librairie Letouzey 
et Aii4). I owe a tirage a part to the author's kindness. 

3 He is inexact in saying (p. 20) that Holder ' signala un nouveau manuscrit de 
VExpositio de Pelage': see below (pp. 203 f.) for the history of the discovery. 

'' Nouv. sir. t. XIV pp. 424 — 450. 

5 t. XIII pp. 225 S. 

6 t. XIV pp. 447 f. 


Viilgate, by referring to the type of text preserved in the Balliol manuscript. 
With the view Lagrange expresses as to the different Old-Latin texts employed 
by the commentators Ambrosiaster, Jerome, Pelagius, I should agree entirely. 
In the Bulletin de Litterature Ecdesiastique de Toulouse^ Father Cavallera 
supports Lagrange's contentions in the Remie Bihlique for 1917 (just mentioned) 
and 1918 (pp. 255 ff.). Cavallera concludes, as I believe rightly, that in 384 
Jerome had issued a revision of the Gospels only, and that such revision of the 
Epistles of St Paul as he made was subsequent to that date. The texts cited 
by Jerome in his celebrated letter to Marcella, epist. 27 § 3 (a.d. 384), have in 
Cavallera's opinion the purpose of direct polemic against those that have been 
angered by Jerome's criticisms. The use of the subjunctive throughout, legant, 
etc., rather excludes the idea of an already existing translation of the Epistles 
by Jerome. 

§ 2. The Text of the Other Parts of Scripture 

Pelagius's notes are short, and the quotations from scripture in 
them are also necessarily short. I have gone carefully through all 
that I could identify and compared the text with the Vulgate 
throughout. Having noted down all the differences, I have collected 
parallels to these from Sabatier and every other source I could 
think of. It is quite evident that Pelagius employed an Old-Latin 
Bible throughout, and the parallels between his quotations and 
other Old-Latin texts ought to be instructive. Considerations of 
space do not permit the complete publication of these data. Without 
them an induction would be unsafe, but it is not necessary to present 
the lists to the reader in full. It will be enough to select the more 
striking examples and to state briefly for each section of scripture 
the relationship with other texts that appears to emerge. It is 
hardly necessary to remind the reader that for all parts of scripture 
except the Gospels our supply of Old-Latin texts is scanty. Owing 
again to the war, I have not been able to avail myself of Pfarrer 
Denk's great work, which under happier conditions would now have 
been completely published. If a critical edition of the Letter to 
Demetrias had been available, the numerous quotations in it would 
also have been at my disposal. This is one of various directions 
in which the present work may be amplified by my successors. 

1 For my reference to this I depend entirely on Father Lagrange in the Revue 
Biblique nouv. ser. t. xvi (1919) p. 283. 


The Heptateuch 

Gen. xlviiii 10 deficiet lugd. Nouat. Cypr. Hil. Hier. Greg.- 
IlUh. Aug. Prom. 
auferetur vg. 
princeps lugd. Nouat. Cypr. Hil. Hier. Amhr. 
Greg.-Illih. Iren.^^^- Aug. Prom. 
sceptrum vg. 
ex lugd. Hier. Ambr. Priscill. Aug. Prom. 

de vg. 
nee lugd. (Nouat.) Hil. (Hier.) Greg.-Ulih. 
(Tren.^^^- Prom.). 
et vg. 
femoribus Ivgd. Nouat. Cypr. Hil. Hier. Ambr. 
Greg.-Illib. Iren.^^^- Aug. Prom. 
femore vg. 

Exod. xxii 20 sacrificans {def. lugd.) m Cypr. Ambst. | Aug. 
qui immolauerit wirceb.^ 
qui immolat vg. 
eradicabitur (def. wirceb. lugd.) m Cypr. Ambst. 
occidetur vg. 

Leu. vii 9 (19) omnis wirceb. lugd. Cypr. Pacian. 
qui fuerit vg. 
manducabit wirceb. lugd. Cypr. Pacian. 
uescetur vg. 
xxvi 12 et inhabitabo in illis (def. lugd.) Ambr.^. 
ambulabo inter eos vg. 
Deut. xxx 6 in nouissimis diebus Cypr. 
om. vg. 
deus {lugd.) Cypr. (Aug.). 

dominus deus tuus vg. 
ad...amanduni Cypr. 
ut diligas vg. 

^ For this, see below under 'Minor Prophets.' 


Deut. xxxii 21 zelauerunt cod. ap. Sabat. 
prouocauerunt vg. 
non deo cod. ap. Sabat. Tert. Avibr. 
eo qui non erat deus vg. 
The agreement between the Lyons Heptateuch and Pelagius's 
text is almost perfect. The further agreement with fourth century 
quotations, and even with Cyprian, is very striking. 

The Historical Books 

1 Regn, xvi 7 in facie m (Cypr.) Hil. Ambst. Hier. {Gaud.). 

ea quae parent vg. 
deus m Cypr. Hil. Ambst. Hier. Gaud. 

dominus vg. 
in corde m (Cypr.) Hil. Ambst. Hier. 

intuetur cor vg. 
3 Regn. iiii (v) 25 (9) sicut Tycon. Ambr. f . 

quasi vg. 
maris Tycon. Ambr. |. 

quae est in litore maris vg. 

The situation here is like that in the Heptateuch. For the 
Historical Books Pelagius evidently employed an average fourth 
century text, which here again has real points of contact with 


xxxi 5 cordis m cod. Sangerm. psalt. Rom. Corb. al. Hil. 
Pacian. Ambr. f Hier. Aug. 
peccati vg. 
Ixxiii 19 animam confitentem cod. Sangerm. Aug. Prom. 
animas confitentes vg. 
Ixxxxiii 12 om. homo Ambr.^. 
habet homo vg. 
cxviiii 7 qui oderant (I) Aug. {non semper), Prosp. 

odientibus (l) psalt. Hebr. cod. Casin.- Hil. Tycon. 
Ambr. ^. 
qui oderunt vg. 

1 On xlviiii 21 ea, see Rahlfs, Septuaginta-Studien 2 Heft (Gottingen, 1907) p. 65. 

2 edidit A. M. Amelli O.S.B. { = Collectanea Biblica Latitia yo\. i) (Romae, 1912). 




The passages here given are not of average character. For 
the most part the quotations made by Pelagius from the Psalms 
agree with that form which is printed in the Vulgate, namely the 
'Gallican' psalter, Jerome's second psalter. The above passages 
are selected out of fourteen phrases which seem to represent a 
slightly older text. 

Sapiential Books 

Prou. V 22 funiculis' -^ug- h codd. Cassian. ^ Fulg. Gildas 
p. 85 codd. (see Thes. s.v. criniculus). 
funibus vg. 
unus quisque Hil.^ Ambr.^ Hier.^ Greg.-IUib. 
Iren}^^- Aug. | Cassian. | Prom. Fulg. Gildas. 
om. vg. 
viiii 8 amabit m Pacian. Hier. Op. imp/. Aug. 
diliget vg. 
xi 26 pretio grauans triticum m. 
qui abscondit frumenta vg. 
populo m. 

populis vg. 
maledictus {m Cypr. Ainhr. |). 
maledicetur vg. 
xiiii 6 quaere[s] Hiei'. Cassian. 
quaerit vg. 
apud malos Hier. Cassian. 

derisor vg. 
inuenies Hier. Cassian. 
inuenit vg. 
xvii 6 gloria patris-. 

corona senum vg. 
eius qui fidelis est totus mundus diuitiarum 
Hier. (Ambr. etc.). 
om. vg. 

^ The variants here are instructive : funibus /Iter, f, restibuslren.'*'- Greg.-Illib., 
liinculis Hil. \ codd., Ambr. | Prom., criniculis Aug. f Gildas, fasciis Hil. f . 

- This is not the only place where Pelagius apparently provides the only existing 
Old-Latin evidence, in all cases unknown to Sabatier : cf. also Prou. xv 13, xviii 17 
(below), xxviiii 27, (lob xxx 25), etc. 

s. p. 11 


Prou. xviii 17 in primordio Aynbst. Ambr. cod. (C.S.E.L. Ixii 73) 
et saep. 
prior vg. 
accusator est Hil. | Amhst. Amhr. 

est accusator vg. 
ut statim cum coeperit aduersarius confun- 
uenit amicus eius et inuestigabit eum vg. 
XX 13 detrahere vi Cypr. Hier. 
somnum vg. 
eradiceris m Cypr. codd.^ Hier. (Paulin.) anon, 
ap. Paulin. 
te egestas opprimat vg. 
XXV 8 uiderint Zosimus (Rome a.d. 417 — 418). 
uiderunt vg. 
loquere Zosimus. 
ne proferas etc. vg. 
xxvi 12 uidi m. 

uidisti vg. 
uirum m. 

hominem vg 
qui sibi sapiens uidebatur. 

sapientem sibi uideri vg. 
habuit m. 

habebit vg. 
quam ille m. 
illo vg. 
Eccl. vii 5 sapientis. 

sapientium vg. 
in domo luctus m Hier. Ambr. f Aug. 
ubi tristitia est vg. 
Cant. V 16 fauces Ambr. f . 
guttur vg. 
dulcedines Ambr. | Hier. 
suauissimum vg. 
lob viiii 28 propter deum. 
om. vg. 

1 True text in Cypr. is extollaris. 


lob xxiiii 23 illi. 

ei vg. 
superbia t. 
superbiarn vg. 
Sap.^ V 2, 3 saiutis insperatae {t). 
insperatae saiutis vg. 

om. vg. 
inter {cod. ap. Sabat.) t Gypr. Lucif. 

intra vg. 
per angustiam t Cypr. Lucif. 
prae aiigustia vg. 
Eccli. iiii 8 sine tristitia om. m cod. ap. Sabat. Ambr. Aug. 
habet vg. 
V 4 accidit mihi Cypr. Lucif. Aug. Ps.- Vigil.-Taps. 
mihi accidit vg. 
est enim altissimus Gypr. Lucif. Ps.-Vigil.-Taps. 
altissimus est enim vg. 
8, 9 neque Aug. Paulin. Oildas. 
et ne vg. 
uenit m cod. ap. Sabat. Gypr. Vict.-Tonn. Oildas. 

ueniet vg. 
eius Gildas. 
illius vg. 
xxi 1 ne Aug. Fulg. Gildas. 
non vg. 
XXXV 11 et. 

om. vg. 
In this section the parallels with m and t will not have escaped 

^ It is perhaps necessary to remind the reader that Sap. Eccli. Bar. 1, 2 Mace, 
are in the Vulgate simply taken over from the Old-Latin. Pelagius's text of Sap. 
Eccli. differs considerably from that in vg, but where parallels to Pelagius's text are 
lacking, I have not thought it necessary to set it down here. 



The Minor Prophets 

Os iiii 5, 6^ nocti adsimilaui Priscill. Hier."^ 
nocte tacere feci vg. 
factus est. 

conticuit vg. 

eo quod vg. 
habens (Priscill.) Hier. 
habuerit vg. 
viii 4 07)1 . ipsi ii^ Filast. {Hier. Gildas^). 
habet vg. 
per IV Cypr. | Hier. Filast. Gildas\. 
ex vg. 
Ion. iiii 11 pecora wirceb.* Ambst. | Hier. 
iumenta vg. 
Zach. ii 8 (12) tangit m Faustin. Ambr. Fulg. 
tetigerit vg. 
tamquam qui tangat {Fulg.). 

tangit vg. 
ipsius m Faustin. 
mei vg. 
Mai. ii 7 exquirant (exquirent) Cypr. (-ent) Ambr. (-unt). 
requirent vg. 
de (1). 

ex vg. 
omnipotentis (I) Cypr. Ambr. Hier. Aug. 
exercituum vg. 
The phenomena here are what we have become accustomed to 
in other sections. 

The Major Prophets 
Esai. i 22 caupones tui Hier. Iren.^^'^- Greg.-Hlib. Max.- 
Taur. Op. -imp/. 

1 Note the remarkable reading resurgens in Os. vi 3. 

2 t in the Prophets is almost pure Vulgate, as also in the Gospels. 

^ For IV, see P. Lehmann, Die Konstanz-Weingartener Propheten-Fragmente... 
(Leiden, 1912). 

* For xoirceb., see Par Palimpsestorum Wirceburgensiuin...'E,. Rauke (Vindob. 


Esai. i 22 miscent uino aquam Iren. (comment) (Macc.- 
(cont.) Taur.). 

uinum tuura mixtum est aqua vg. 
vii 9 nizi Tert.^ Priscill. Ambst. Greg. -Illih. Aug. ^. 
si non vg. 
nee Tert. \ Amhr. neque Gypr. | Amhst. 

non vg. 
intellegetis Tert. f Gypr. | {ap. Hier.) Amhst. 
Priscill. Greg.-Illib. Ambr. Aug .^Prom.Fulg . 
permanebitis vg. 
1 1 peccatis Ambr. f Hier. Rufin. 
in iniquitatibus vg. 
uenundati Hier. Rufin. 
uenditi vg. 
liii 7 uictimam Tert. Gypr. Ambr. Hier. Iren.^^^ Gaud. 
occisionem vg. 
duetus est Tert. Gypr.'^°^'^- Hil. Amhst. Amhr, 
Hier. Iren.^''^- Aiig. Gaud. Prom. Vigil.-Taps. 
ducetur vg. 
Ixvi 2 humilem m Gypr. | Nouat. Hil. (Amhst.) Pacian. 
Hier. Ambr. saep. Iren.^^^- Aug. f Gassian. 
pauperculum vg. 
quietum m Gypr. | Nouat. Hil. ^ Pacian. Hier. 
Ambr. saep. Iren.^^^- Aug.§ Gassian. 
contritum spiritu vg. 
uerba m Nouat. Hil. | Amhst. Pacian. Ainhr. 
saep. Aug.^. 
sermones vg. 
Hierem. iii 15 secundum Gypr. | Lucif. | Aug. 
iuxta vg. 
cum disciplina Gypr. | Lucif. f Aug. 
scientia et doctrina vg. 
iiii 3, 4 innouate m Tert. Greg.-Illib. 
nouate vg. 
nouamen Tert. f . 

nouale vg. 
ne seminaueritis Gypr. 
nolite serere vg. 


Hierem. iiii 3, 4 in m Tert. Gypr. Macroh} 
(cont.) super vg, 

circumcidite uos Cyp7\ Amhst. Macrob. 

circumcidimini vg. 
deo uestro m Tert. Cypr. Macroh. Hier. 

domino vg. 
circumcidite m Tert. Cypr. Amhst. Macroh. G-reg.- 
auferte vg. 
praeputium m Tert. 4 Cypr. Amhst. Macroh. 

praeputia vg. 
cordis uestri m Tert. Cypr. Amhst. f Macroh. 
cordium iiestrorum vg. 
viiii 23 sua sapientia 7?i| < Zwa/. |. 
sapientia sua vg. 
neque t (Hil.). 

et non glorietur vg. 
neque (alt.) (Hil.) Lucif. 
et non glorietur vg. 
xvii 16 concupiui Hil. Aug. 

desideraui vg. 
xxiii 24- nonne wirceh. Cypr. ^Hil.Hier. (semel)Nicet. Fulg. 
numquid non vg. 
Ezech. xviii 20 ipsum m lulian.-Aecl. Gildas. 
eum vg. 
xxxiii 11 peccatoris (w illegible) Tert. (Cypr.) Pacian. 
Ambr. Hier. Aug. Gildas^. 
impii vg. 
Dan. ii 47 ipse est deus Amhst. 
deus deorum est vg. 
viiii 23 concupiscentiarum^ 
desideriorum vg. 
The evidence for this section of the Bible is of the same nature 
as that which we have found for all the rest of the Old Testament. 

1 I.e. the Donatist, in the opinion of Harnack and Morin the true author of the 
Pseudo-Cyprianic Be Singidaritate Clericorum. 

• On this important word see Konsch, Itala u. Vulgata p. 49, Thes. Ling. Lat. 
iv 102 74 ff. It is real Old-Latin. 


The Gospels 

If my numeration be right, there are, for Matthew, 23 agree- 
ments with A, 22 agreements with c, 20 agreements with g, 19 agree- 
ments with a, 19 agreements with 6, 15 agreements with c?, 15 
agreements with q, 14 agreements with e, 12 agreements withji^"-, 
9 agreements with /, 8 agreements with m, 7 agreements with k, 

4 agreements with ff, 4 agreements with I, 3 agreements with g^, 
2 agreements with t, against the Vulgate. In estimating the value 
of these figures, one must remember that not all these Old-Latin 
texts contain the whole Gospel. In fact c,f,ff, g, g^, I alone seem 
to be complete. Pelagius obviously employed a real Old-Latin text 
in Matthew, and if we could say that h, so far as Matthew is con- 
cerned, is practically what Pelagius had before him, it would be a 
point gained. If we similarly compare the patristic authorities, we 
shall find that Pelagius agrees against the Vulgate with Augustine 
17 times, Ambrose 15 times, Cyprian 12 times, the Opus Imper- 
fectura 10 times, Hilary 9 times, Ambrosiaster 8 times, Tertullian 
7 times, the Latin Irenaeus 6 times, Gildas 6 times. Chromatins 

5 times. Here again it must be remembered that the Opus Im- 
perfectum, Hilary and Chromatins are commentators on Matthew, 
also that Augustine's works are the most voluminous and he quotes 
very copiously from the Gospels. I attach most significance to the 
15 agreements with Ambrose and the 8 agreements with Ambro- 
siaster. It is remarkable that we have also found Ambrose to have 
important points of contact with the codex of the Pauline Epistles 
used by Pelagius. Nor will the agreements with Gildas escape the 
reader's notice. 

The Marcan agreements against the Vulgate are naturally very 
few, and do not merit mention, with the one exception of 

Mark xvi 17 credentes c {q) Amhr.^ Prom. 
eos qui crediderint vg. 
haec signa. 

signa...haec vg. 

In Luke there are, against the Vulgate, 7 agreements with a, 

6 with d, 5 with e, 4 with c, 3 with I, 3 with r. If we could go so 
far as to say that Pelagius employed the a text for Luke, it would 


be an interesting point of contact between him and Jerome ^ One 
or two of the citations are of sufficient interest to be presented in 

Luke vii 47 cui plus dimittitur \ , \ diligit. 

[plus j 

Gypr. codd. (plus in both places), Iren. (plus in 
both places), Ambr. | (plus in both places in 
one citation, multum in the first place in 
another). The Armenian version has also this 
clause, which is wanting to the Vulgate, as it 
is to every extant Old-Latin MS. If Pelagius 
found it in his copy of Luke, then it was dif- 
ferent from a. It is however possible that he 
got it from Cyprian, As he was himself the 
author of a book of extracts, he must have been 
quite well acquainted with Cyprian's Testi- 
Tiionia. Observe again the contact with 

XX 36 del sunt. 

sunt dei vg. 
Note that this clause is absent from most of the 
Old-Latin authorities. 
38 non est deus a Cypr. codd. Aug. 
deus non est vg. 
illi uiuunt adeir Cypr. Pacian. Aug. 
uiuunt ei vg. 
xxi 34 om. forte f Amhst. Amhr. Adim. ap. Aug. 
habet vg. 

In the Fourth Gospel there are 9 agreements with a, 8 agree- 
ments with h, 7 agreements Avith e, 7 agreements with r, 5 agree- 
ments with c, 5 agreements with /, 4 agreements with d, 4 agree- 
ments with ff'\ 4 agreements with m, against the Vulgate. This 
evidence appears to point to the Pelagian John as a fourth century 

1 Cf. the writer in Journal of Theological Studies vol. xii (1910—11) pp. 583 ff., 
and Hoskier, The Golden Latin Gospels (New York, 1910) pp. xxix, cxiv. The text 
of Luke xvi 3 — 14 in Hier. cjnst. 121 § 6 differs greatly from a, however. 

^ This Agraphon finds no place in Eesch's Agrapha. 


Old-Latin type of text. One or two verses will illustrate its cha- 
racter and relationship. 
John viiii 39 ego 07n. 

hahet vg. 
ueni in hunc mundiun abcfqr Ambst. f Hier. 
in hunc mundum ueni vg. 
xii 35 ne a el m r Arnob.-Iun. 
ut non vg. 
xvii 3 unum a Nouat. ^ Aug. Arnob.-Iun. 
solum vg. 
24 isti Ambr. f A mbst. HI • 
illi vg. 
This last is a pretty example, pointing again in the direction 
of Ambrose. 

The quotations from Acts are of a somewhat unsatisfactory 
meagreness, but we find 9 agreements with d, 8 with p, 7 with e, 
5 with gig, 4 with c, and 4 with dem, against the Vulgate. Here 
the most significant fact is to be found in the agreements with p, 
which is Old-Latin for just half the Book of Acts. Out of 18 possible 
agreements there are thus 8 to be found. The relationship is not, 
therefore, very close. We cannot identify any existing Old-Latin 
MS as certainly the type employed by Pelagius, Of the patristic 
parallels, 7 are to be found with Augustine, and 4 with each of the 
following, namely Gregory of Elvira, Petilian, Ambrosiaster and 
Isidore : while in the case of each of the following parallels, Pseudo- 
Prosper, Latin Irenaeus and Niceta, Pelagius shares 3 instances 
against the Vulgate. The evidence suggests that the British text 
was related to those used in Africa and in Spain rather than any 
others. There are two parallels with Gildas. 

Acts ii 3 uisae sunt d Cypr. Greg.-Illib. Ambr. (ps. 38 
p. 196) Aug. Petil. Prom. Vigil. 
apparuerunt vg. 
eis {d) e Aug.^ Petil. Prom. Vigil. 

illis vg. 

dispertitae vg. 



Acts ii 3 

quasi Greg.-Illih. Aug. f Prom. 


tamquam vg. 

super del Greg.-IUih. Aug. Petil. 

supra vg. 

iii 17 

hoc CT (both Spanish MSS). 

oin. vg. 

iiii 31 

omni uolenti credere de J) etc. Iren. (cf. uni 

cuique hominum uolenti credere lectionar. 

Schlettst. [Morin, Etudes I p. 449J). 

oni. vg. 


umim 1 c demgigp DW al. Ambr. ps. 47 p. 351, 


om. vg. 


distrahentes f . 

uendentes vg. 



ponebant vg. 

Ad dp Aug. Petil. 

ante vg. 

V 41 

apostoli dp Ambst. 

illi vg. 

quia dp. 

quoniam vg. 

domini (|) (e) D0. 

lesu vg. 


om. vg. 

xiii 2 

segregate (|) c dem gig m p etc. AKMOV W Lucif. 

Ambst. Nicet. Isid. etc. 

separate vg. 

Paulum (1) mp Ambr. | Nicet. Isid. 

Saulum vg. 

ad d gig Ambst. Nicet. Isid. 

in vg. 

elegi Ps.-Vigil. Isid. 

adsumsi vg. 

XX 26—28 

ab omnium sanguine Gildas. 

a sanguine omnium vg. 


Acts XX 26 — 28 uobis omne consilium dei e, Gildas (mysterium). 
(cont.) omne consilium dei uobis vg. 

uidete gregem. 

adtendite...gregi vg. 

adquisiuit vg. 
episcopos ordinauit. 
posuit episcopos vg. 
XX vi 19 caelesti uisioni c dem (e) gig p (D)R*W etc. 
caelestis uisionis vg. 
These quotations may suffice to show the character of the Acts 
text used by Pelagius. 

The Epistle to the Hebrews 

The agreements with other texts are very scanty. Pelagius 
regarded Hebrews as Pauline, but hardly as canonicals The text 
he used appears to have been somewhat different from any other 
known text; for out of thirteen differences from the Vulgate, nine 
are unparalleled. Let me give here those that are paralleled: 

Heb. i 3 imago expressa (figura expressa D) (imago d t 
Amhr. ps. 38 p. 202) Sedul. 
figura vg. 
vi 4 inluminati sunt d Tert. Amhr. 

sunt inluminati vg. 
X 36 repromissionem Z d. 
promissionem vg. 
xii 25 recusantes om. d. 
habet vg. 
The situation might be reconciled with the view that Pelagius 
used a text like d, but quoted frequently from memory. It is 
interesting to note that he began the new sentence with defici- 
entes (xii 3): to this I know no parallel. 

Canonical Epistles 

The quotations from James are hardly worth noting. Of six 
differences from the Vulgate, four are unparalleled. But in iii 14 
^ Cf. De Bruyne, Rev. Biblique for 1915, p. 373. 


we find si enim with (ff) against quod si of vg, and in iiii 4 deo with 
DF against dei of vg. 

In First Peter the following may be noted: 
1 Petr. i 19 inmaculati et incontaminati A. 
incontaminati et inmaculati vg. 
ii 1, 2 simulationem t Gildas. 
simulationes vg. 
tamquam Aug. 

sicut vg. 
rationale Hier. 4 Prom. 
rationabile vg. 
7 incredulis Hie7\ 

non credentibus vg. 
23 qiii percussus est et non percussit (Amhr. f 
om. vg.^ 
iii 9 neque D (Aug.). 
uel vg. 
Special attention must be called to ii 23. This point of contact 
with Ambrose and his fellow is important in view of what has been 
said about Pelagius's Pauline text in § 1. 

For Second Peter only one quotation is of moment: 
2 Petr. iii 9 promissi(?) G m. 
promissis vg. 
aliquem A m Aug. Fulg. 

aliquos vg. 
conuerti Aug. Fulg. 
reuerti vg. 
Here again, as in Acts, we are taken to Africa and Spain, but 
of course it must be remembered that the total evidence for the 
use of Second Peter is scanty. 

Pelagius was rather fond of First John. A few examples of 
readings may be given fi'om that epistle: 

1 loh. ii 6 Christo (|) Cypr. \ Hier. | Aug. | Fulg. 
ipso vg. 

1 Cf . C. H. Turner, Ecclesiae Occidentalis Monumenta fasc. i (1) (Oxou. 1899) pp. 
xvi, 33 ; Sanday, Novum Testamentum S. Irenaei (Oxford, 1922) p. 189. 


1 loh. ii 6 quo modo (|) Cypr. |. 
(cont.) sicut vg. 

17 fecerit Cypr.^ Lucif. Aug.^ Gelas. 

facit vg. 
21 d.e m Iren}^^- 
ex vg. 
iii 16 quia hr Ambst. Aug. 
quoniam vg. 
debemus ergo et nos. 

et nos debemus vg. 
nostris m. 
om. vg. 
iiii 20 si fratrem quern uides, non diligis (Ambr.) 
Aug. I. 
qui non diligit fratrem suum quem uidet vg. 
uides Ambr. Aug. I. 

uidet vg. 
potes Ambr. 
potest vg. 
The text does not seem to have differed much from what would 
have been found in Africa in the fourth century. 


Only one reading need be mentioned. It appears that in vi 8 
Pelagius read: 

peccatum et mors (vg and all other authorities mors simply). 

This is important, unless we are to regard it as a persistent 
lapse of memory on his part. 



The question of the sources of any ancient work is complicated 
by the loss of so much of the old literature. The study of the sources 
of commentaries on the Pauline Epistles is above all hindered by 
the disappearance of the greater part of Origen's exegetical work. 
He was in a very real sense the father of Biblical exegesis, and every 
investigation of this kind suffers from that loss. In the present case 
the difficulty of the task is accentuated by the very brevity of the 
notes, which are the concentrated essence of much reading and 
meditation. There has been no serious attempt to analyse the 
Pseudo-Jerome commentary. Nothing has been written on the 
subject, so far as I know, except the few paragraphs of Richard 
Simon^ and the list of parallels between Pelagius and the 
latinised Theodore of Mopsuestia drawn up by the late Professor 

This neglect has not been altogether disadvantageous. Until 
the text of Pseudo-Jerome was purified of its thousands of textual 
corruptions, and was seen to fall apart into at least two portions, 
one written by Pelagius, the other by a Pelagian, it would have 
been premature and confusing to formulate any conclusions with 
regard to the sources. When it fell to me to make the necessary 
discoveries with regard to the form and character of the Pseudo- 
Jerome, I resolved to study also the possible Greek and Latin 
sources of at least that part which comes from Pelagius himself. 
But as the purely textual and linguistic work grew in my hands, 
and it became evident that the further task would delay publica- 
tion for several years, I was glad to avail myself of the willingness 

^ Histoire Critique des principaux Commentateurs du Nouveau Testament (Rotter- 
dam, 1693) pp. 236 ff. 

2 Theodori Episcopi Mopstiesteni in epistolas B. Pauli Covimentarii -vol. i (Cambr. 
1880) pp. Ixxiv ff . 


of my former pupil Dr Alfred J. Smithy to undertake this part of 
the investigation. In the first instance he has confined himself to 
the most important part of the commentary, that on the Epistle 
to the Romans, and to the Latin sources of this part; but he hopes 
to complete the task later. Such work on the sources of the com- 
mentary as I have been able to perform myself, has been unsys- 
tematic, but it has been here incorporated for any worth it may 
happen to have. To Dr Smith will always belong the credit of being 
the first serious investigator into the sources of the Pelagian ex- 

Our best plan would seem to be to take, in chronological order, 
the Greek and Latin works which seem to have a close connexion 
at points with the commentary of Pelagius. An exception is made 
in the case of Origen, for it would appear that Pelagius knew, not 
the Greek Origen, but only the Latin of Rufinus's " translation ^" 

Before proceeding, however, to these details, it is necessary to 
remark that Pelagius makes no reference hy name to any authority 
employed by him. In this respect he is only following the ancient 
practice. He does, however, not infrequently refer to the opinions 
of others by the vague terms quidam, multi, alii, diuersi, and it will 
help other scholars to supplement the serious defects of the present 
chapter, if I give here a list of the passages where these vague 
expressions occur. I am not without hope that those that really 
know patristic literature, as a mere Latinist can hardly be expected 
to know it, will be able to answer in many, if not in all cases, who 
the writers referred to are^. It seems probable that these authori- 
ties were for the most part recent, perhaps also for the most part 

1 M.A., D.Litt. (Aberdeen), at that time Classical Master in Robert Gordon's 
College, Aberdeen, Lieutenant, R.N.V.R. Dr Smith's articles are published in the 
Journal of Theological Studies vol. xix (1917 — 18) pp. 162 — 230; vol. xx (1918 — 19) 
pp. 55—65, 127—177. 

2 Smith, in J.T.S. vol. xx, pp. 127 ff. 

^ Dr Smith has given the answer in certain cases. As the result of a somewhat 
cursory study I have formed the opinion that TertulMan, Cyprian, Novatian, 
Lactantius and Hilary (on the Psalms) are not among the quidam referred to except 
perhaps in these passages: Tertullian, de pudic. 13 (C.S.E.L. xx p. 245, Oehler, 1. 1 
p. 819) colaphis . . .cohibebatur per dolorem ut aiunt auriculae uel capitis, may be al- 
luded to on 2 Cor. xii 7 quidam dicunt eum frequenti dolore capitis laborasse ; Cyprian, 
epist. 5i § 3 (Hartel, pp. 622 f.) may be among those alluded to on 2 Tim. ii 20. 


Qiddam are referred to in the notes on: 

Rom. ii 21 Rom. x 5 1 Cor. xv 28 Gal. iii 19 

iii 28 xi 20 (also multi, v 12 

iiii 7 26 alii) Eph. i 10 

(also alii) xiiii 2 2 Cor. ii 15 (also multi, 

V 14 15 iii 6 alii) 

viii 3 1 Cor. ii 9 v 13 iii 18 

19 viii 1 vii 11 v 31 
viiii 16 viiii 22 viii 22 vi 5 

20 bis xi 21 viiii 1, 2 Col. ii 14 
(21) 27 (also alii) 2 Tim. ii 20 
( 26) xiiii 19 xii 7 

Multi, diuersi are referred to in the notes on : 

Rom. viii 19 Eph. i 10 

viiii 17 ii~ 2 

1 Cor. XV 28 Phil, ii 5 


It would have been very strange if Pelagius had made no use 
of the work of his remarkable predecessor, who in the period 365 — 
380 issued a set of commentaries on the thirteen epistles of Paul. 
Who this predecessor was may be said to be still a subject of debate, 
even after the intensive study of his commentary which the last 
twenty years have seen. The most clamant need now is a reliable 
edition of the various ancient forms of the commentary. It Avould 
appear that the commentary on Romans was in the first instance 
issued by itself, possibly while Hilary of Poitiers was still alive. 
It was certainly known both in Africa and in Ireland under the 
name "Hilarius." At a later date the commentaries on the other 
epistles were issued: to these the name Hilarius was never attached, 
and the whole work, comprising an enlarged form of the commentary 
on Romans, circulated both anonymously and under the name of 
Ambrose in the succeeding centuries. The manuscript tradition is 
involved in various ways which it must be left for Father Brewer, 
the future Vienna editor, to set forth. 


Pelagius may have been influenced by several motives to produce 
a new and complete Latin commentary on the Epistles. In the first 
place, the considerable length of the Ambrosiaster commentary, 
which probably filled two codices of the size fashionable in the fourth 
century, made a shorter exposition, which could be comprised in 
one codex, extremely desirable. In the second place, Ambrosiaster 
was a pronouncedly Western expositor, with a hostile attitude to 
Greek writers and manuscripts, while Pelagius seems to have known 
and valued Greek Christian writings. Quite apart from this differ- 
ence of attitude, Pelagius definitely disagreed with certain opinions 
expressed in the Ambrosiaster commentary. We shall see, however, 
that he could not help using a work of such originality and import- 
ance. From Dr Smith's exhaustive account of the relations of the 
two commentaries in the Epistle to the Romans, I will borrow only 
the most striking coincidences. 

Rom. i 8. Pelagius: '(Paulus) prudenter laudat, ut prouocet 
ad profectum.' Ambst.: 'hortatur ad profectum.' 

Rom. i 22, 23. Pelagius: 'nee in hominis solum, sed in imaginis 
hominis' (sc. similitudinera). Ambst.: 'ut...non hominibus, sed 
quod peius est et inexcusabile crimen, in similitudinem hominum 

*P.'s comm. on Rom. i shew throughout traces of the influence 
of Ambrosiaster^' 

Rom. ii 14 — 16. Pel: 'conscientia et cogitationes nostrae erunt 
in die iudicii ante oculos nostros.' Ambst.: 'gentes...per conscien- 
tiam iudicabuntur, si credere noluerint.' 

Rom. ii 25. Pel.: 'quo modo ergo nihil est (sc. circumcisio), si 
prodest? sed prodest tempore suo. signum prodest, si iustitia cuius 
est signaculum, adsit; ceterum sine ilia superfluum erit.' Ambst.: 
'potest dici e contra: "si prodest circumcisio, cur praetermittitur? 
sed tunc prodest si legem obserues'" etc. 

Rom. iiii 4. Pel.: 'debitoris enim est facere quae iubentur, et 
nisi paruerit, damnatur.' Ambst.: 'debitoris est enim facere legem, 
quia necessitas imposita est per legem, ut uelit nolit faciat legem, 
ne damnetur.' 

Rom. iiii 5. Pel.: 'conuertentem^ impium per solam fidem 

1 Smith, J.r.S. XIX p. 175. 

^ Here as present participle of deponent conuertor. 

S, P. 12 


iustificat deus.' Ambst.: 'impius per solam fidem iustificatur apud 
deum.' Pel.: 'quo proposuit gratis per solam fidem peccata di- 
mittere.' Ambst.: 'sic decretum dicit a deo, ut cessante lege sola 
fides gratiae dei proficeret ad salutem.' 

Rom. V 15. Pel.: 'iniustum esse dicentes ut hodie nata anima 
non ex massa Adae tam antiquum peecatum portet alienum, dicunt' 
etc. Ambst. on Rom. v 12: 'manifestum itaque est in Adam omnes 
peccasse quasi in massa \' Note especially this extraordinary use 
of massa, which Augustine quotes from Hilarius (Ambrosiaster) in 
his Contra duas Epistidas Pelagianorum iv 4 § 7. 

Rom. vii 8 — 10. Pel. : 'hie peecatum diabolum uidetur appellare, 
sicut et in Apocalypsi nominatur, scilicet ut auctor peccati.' Ambst. 
on Rom. vii 4 'peecatum quod est diabolus,' and especially on 
Rom. vii 11 'peecatum hoc loco diabolum intellege, qui auctor 
peccati est-.' Pel. 'quod {sc. mandatum) custoditum proficiebat 
ad uitam, neglectum duxit ad mortem.' Ambst. 'quia (lex) oboe- 
dientibus proficit ad uitam.' 

Rom. vii 18. Pel.: 'non dixit {sc. Paulus): "non est caro mea 
bona.'" Ambst.: 'non dicit, sicut quibusdam uidetur, carnem 

Rom. vii 22. Pel. on Rom. v 15: 'si anima non est ex traduce, 
sed sola caro, ipsa tantum habet traducem peccati et ipsa sola 
poenam meretur'; Pel. on Rom. vii 17: 'habitat {sc. in me peecatum) 
quasi hospes et quasi aliud in alio, non quasi unum, ut accidens 
scilicet, non naturale.' Ambst.: 'non in animo habitat peecatum, 
sed in carne quae est ex origine carnis peccati, et per traducem fit 
omnis caro peccati. si enim anima de traduce esset et ipsa, et in 
ipsa habitaret peecatum, quia anima Adae magis peccauit quam 

corpus in carne ergo habitat peecatum quasi ad ianuas animae, 

ut non illam permittat ire quo uult: in anima autem si habitaret, 

^ This important coincidence between Ambrosiaster and Pelagius has not been 
alluded to by Dr Smith, because it is a coincidence between notes on different verses, 
but it has not escaped E. Buonaiuti, La grnesi della dottrina agostiniana intorno al 
peccato originale (Roma, 1916) p. 15, n. 4, and was noted by the present writer a 
number of years ago. Cf. for this word Paris MS. 653 on Rom. xv 8, Aug. Ep. 186 § 4, 
Ambst. Rom. viiii 21, the proem to the De indiaatione cordis Pharaonis, published 
by Morin in Rev. Bened. xxvi p. 179 dtias massas hiimanae naturae, boiiam et malam, 
a deo esse factas. 

2 On a development of the text of Apoc. vi 8, personal to Pelagius himself, see 
chap, iv p. 173. 


numquam se cognosceret homo: nunc autem cognoscit se et con- 
delectatur legi dei.' 

Rom. vii 24, 25. Pel.: 'homo carnalis duplex est quodam modo 
et in semet ipso diuisus.' Ambst. : 'duplex est homo carne con- 
cretus et animo.' 

Rom. viii 2. Pel.: 'quae (lex) peccatoribus data est et mortificat 
peccatores. Sine: Ab ea lege, quam supra dixerat in membris esse.' 
Ambst.: 'lex peccati est quam in membris dicit habitare, quae 
suadere nititur aduersa: lex uero mortis Moysi lex est, quia morti- 
ficat peccatores.' 

Rom. viii 20, 21. Pel.: 'uanitas est omne quod quandoque 
finitur.' Ambst.: 'quid est ergo in quo uanitati subiecta est, nisi 
quia quae generat caduca sunt' etc. 

Rom, viii 24, 25. Pel.: 'fides per pa tientiamgrandis est praemii.' 
Ambst.: 'credentes praemiis adficiendi sunt.' 

Rom. viii 28 — 30. Pel.: quos praesciuit:...'quos praesciuerat 
credituros.' Ambst. on Rom. viii 21; 'quos scit deus credituros.' 
There are many parallels to these clauses in both writers. It is not 
too much to say that Pelagius, like Augustine, got his doctrine of 
foreknowledge from Ambrosiaster^ Pel.: ' tempore 
est.' Ambst.: 'si...eligantur ad tempus.' 

Rom. viii 35 — 37. Pel.: 'post tanta et tarn praeclara beneficia.' 
Ambst.: 'Christi qui nobis tam magna et innumera praestitit 

Rom. viii 38, 39. Pel.: 'pro certo confido quia nee si mihi quis 
mortem minetur nee si uitam promittat nee si se angelum dicat a 
domino destinatum nee si angelorura principem mentiatur nee si 
in praesenti honorem conferat neque sipolliceaturgloriam futurorum 
neque si uirtutes operetur nee si caelum promittat et inferno 
deterreat uel profunditate scientiae suadere conetur, umquam nos 
poterit a Christi caritate secernere.' Ambst.: 'quid enim si mors 
inlata fuerit, nonne lucrum est maximum. . . ? neque si praesens uita 
promissa nobis fuerit dignitate quidem si se angelus 
nobis ostendat ad seducendos nos... neque si uirtus ab aliquo facta 
fuerit... neque si in altitudinem se nobis ostendat (sc. satanas)... 
neque si per fantasiam...profundum nobis ostendat horrore miran- 

1 Cf. Smith, p. 163 as well as pp. 201 f., and his note 13 on Augustine, where 
Aug.'s passages on foreknowledge are collected, and chap, iii above, p. 70. 



dum, qua territi forte succumbamus illi...neque si futura nobis 
spondeat ' 

Rom. viiii 1 — o. Pel.: 'contra ludaeos acturus primum illis 
satistacit non se odii causa haec dicere, sed amoris, eo quod doleat 
illos Christo non credere.' Ambst.: 'quoniam superius contra 
ludaeos loqui uidetur...nunc ut uotum suum et adfectum circa eos 
ostendat..,ac per hoc dolet genus suum... quia incredulitate sua 
hoc.beneficio se priuarunt.' 

Rom. viiii 14, 15. Pel.: 'hoc recto sensu ita intellegitur: illius 
miserebor, quern ita praesciui posse misericordiam promereri, ut iam 
tunc illius sim misertus.' Ambst.: 'eius miserebor, cui praescius 
eram quod misericordiam daturus essem, sciens conuersurum ilium 
(cf. Pel. on vv. 11, 12 si conuerti uoluerit), et permansurum apud 

Rom. viiii 17. Pel.: 'tale est hoc quod in Pharaone gestum est, 
quale si medicus de cruciatu iam damnati rei multis inueniat sani- 
tatem, causas inquirendo morborum....' Ambst.: 'hoc etiam genere 
antiqui medici in hominibus morte dignis uel mortis sententiam 
consecutis requirebant quo modo prodessent uiuis, quae in homine 
latebant apertis, ut his cognoscerent causas aegritudinis et poena 
morientis proficeret ad salutem uiuentis.' 

Rom. X 4. Pel.: 'talis est qui Christum credidit die qua credidit, 
quasi qui uniuersam legem impleuerit.' Ambst.: 'hoc dicit quia 
perfectionem legis habet qui credit in Christum.' 

Rom. xi 1, 2. Pel.: 'illam plebem non reppulit quam praesciit 
esse credituram.' Ambst.: 'quos iideles sibi futuros sciuit... 
quam praesciuit deus saluandam...hi quos praesciuit deus cre- 

Rom. xii 13. Both Pel. and Ambst. mention Abraham and Lot 
as dispensers of hospitality. 

Rom. xii 15. Pel.: 'ut siquid patitur unum membrum, con- 
patiantur omnia membra (1 Cor. xii 26).' Ambst.: 'hoc est quod 
alio loco dicit: "siquid patitur unum membrum, conpatiuntur omnia 
membra." ' Pel. : ' cum uiderem homines in necessitate (lob xxx 25) ? ' 
Ambst.: 'cum enim quis fideli solacio^ est in necessitate.' 

' See above, on Eom. viii 28 — 30. 

- Fideli here is masc, while solacio is predicative dative (cf. Eoby's Grammar 
vol. II p. liva). 


Rom. xiii 13. Pel.: 'comissatio est mensae collatio.' Ambst.: 
'(conuiuia) quae...collatione omnium celebrantur.' 

Rom. xiiii 1 — 4. Pel. disagrees with the opinion of 'quidam' 
who think that Paul is here referring to the Jews. As this latter 
view is found in Ambst., he is clearly one of the 'quidam.' 

Rom. XV 13 f Pel.: 'bonus doctor laudando prouocat ad pro- 
fectum, ut erubescerent tales non esse quales ab apostolo esse cre- 
debantur.' Ambst.: 'per laudem... prouocat eos ad meliorem et 
intellectum et uitam. qui enim uidet se laudari, data opera elaborat 
ut uera sint quae dicuntur.' 

Rom. XV 20. Both commentaries refer to pseudo-apostoli. The 
frequency of such references is almost certainly due to the influence 
of the Marcionite prologues \ 

Rom. XV 22. Pel.: 'exposuit illud quod in capite dixerat: "et 
prohibitus sum usque adhuc (i 13)."' Ambst.: 'quod in capite 
epistulae memorat dicens: "quia saepe," inquit, "proposui uenire ad 
uos et prohibitus sum usque adhuc.'" 

Rom. xvi 4. Pel.: 'se periculis obiecerunt.' Ambst.: 'ut peri- 
cula pati pro eo non abnuerent^.' 

1 Cor. iii 17. Pel.: 'suum corpus peccando.' Ambst.: 'turpiter 
uiuentes corpora sua uiolando corruperant*.' 

1 Cor. viiii 5. Both Pel. and Ambst. are definitely of opinion 
that mulieres (muUerem) does not mean 'wives.' Pel.: 'mulieres, 
quae necessaria de suis facultatibus ministrarent.' Ambst.: 'mu- 
lieres. . .sequebantur apostolos ministrantes eis et sumptus et seruitia, 
sicut et saluatorem sunt secutae ministrantes ei de facultatibus 
suis.' It is only right, however, to point out that the same view 
is taken also by Jerome, adii. louin. i 26 (vol. ii- p. 277 D Vail.): 
'non uxores debere intellegi, sed eas, ut diximus {=adu. Heluid. 
11?), quae de sua substantia ministrabant.' 

1 Cor. viiii 20. Pel: 'potest et ita intellegi quod Samaritanos 
dixerit "esse sub lege," quia legem tantum Moysi uidentur accipere.' 

1 Dr Smith is not responsible for this statement. 

2 From this point onward the brief notes are due to the author, as Dr Smith 
has confined his attention to ' Romans. ' 

3 I have a note to the effect that Pelag. on 1 Cor. v 8 in septem diebus—reuol- 
uuntur is paralleled in Ambst., but the nearest parallels I can now find are Quaest. 84 
§ 2, 95 §2; Anon. in. Math. (J.T.S. v (1903—4) p. 236). 


Ambst.: '"hi qui sub lege sunt" Samaritani noscuntur: legem enim 
solam accipiunt, id est quinque libros Moysi\' 

1 Cor. xi 7. Pel.: 'uir ad imaginem dei factus est et idcirco 
liber est: mulier uero ad uiri similitudinem est formata; unde iube- 
tur esse subiecta.' Ambst. on xi 8 — 10: 'mulier ergo idcirco debet 
uelare caput quia non est imago dei, sed ut ostendatur subiecta' etc. 

1 Cor. xi 10. Pel.: 'uelamen signum potestatis esse declarat.' 
Ambst.: 'potestatem uelamen significauit.' 

Gal. arg. Pel.: 'Galatas...pseudo-apostoli hac astutia sub- 
uertebant.' Ambst.: 'Galatae uel ei qui eos subuertebant.' 

Eph. vi 5. Pel. : ' non uenit Christus condiciones mutare : nam 
seruitium non natura dedit, sed captiuitas fecit, neque ex maledicto 
Cham, ut quidam putant, cum ex eo reges legamus esse generatos. 
maledictum uero illud in filiis Cham impletum est, qui Istrahel 
sunt filiis subiugati.' Ambst. on Col. iiii 1: 'dum alter alterius 
fines inuadit, tunc captiuos ducit ingenuos....denique peccati causa 
Cham seruus audiuit: "maledictuspuer Chanaan; seruus seruorum 
erit fi:'atribus suis-.'" 

Phil, iii 2. Pel.: 'hinc contra pseudo-apostolos agit Christi 
euangelium lacerantes.' Ambst.: 'primum oblatrant, dehinc morsu 
saeuo male operantur in carnem.' 

Phil, iii 19. Pel,: 'in circumcisione uerecundi membri.' Ambst.: 
'gloriantes in pudendis circumcisis^.' 

Col. ii 21. Pel. and Ambst., with Ambr., regard these prohibi- 
tions as the Apostle's own: Pel.: '"ne tetigeritis neque gustaueritis 
neque contractaueritis (A)," illo tactu et gustu et contrectatione 
quo hi(?) qui in concupiscentiis abutuntur et diligunt pro aeternis.' 
Ambst.: 'prohibet omni genere ab spe mundanorum, quia inanis 
est' etc. 

1 Tim. V 19. Pel.: 'iniustum est etiam aduersus laicum accu- 
sationem recipere, cum hoc nee saeculares indices faciant: quanto 
raagis aduersus domini sacerdotem!' Ambst.: 'quoniam huius or- 
^ It is to be noted that this view is found already in Origen (Cramer, Catena v 
p. 178), as I learn from Harnack, Texte und Untersuchungen Bd. xlii (4) [1919] 
p. 86, n. 3. 

- On this passage of Ambst. cf. A. J. Carlyle, A History of Mediaeval Political 
Theory in the West vol. i (1903) p. 113 : the scripture quotation is omitted by some 

•^ In the case of Ambst. the scripture text was gloria in pudendis illorum, which 
is nearer to the comment, but there is no trace of this reading in Pelagian MSS. 


dinis sublimis honor est (huiusmodi enim uicarii Christi sunt), 
idcirco non facile de hac persona accusatio debet adraitti.' 

2 Tim. ii 20. Pel. disagrees with 'quidam/ who define the 
'magna domus' as the Church. Ambst. is one of these. 

2 Tim. iiii 20. Pel: 'hinc probatur quia non propter sanitatem 
carnalem tantum apostoli curabant, sed ut etiam signa monstrarent, 
quia hie suum discipulum non curauit.' Ambst.: 'quid est tamen ut 
praesente apostolo qui mortuos excitabat, Trophimus infirmaretur? 
sed signa propter incredulos facta sunt^' etc. 


A few parallels with passages in the works of Jerome written 
prior to our commentary, are here set down. There is no antecedent 
improbability, but rather the reverse, that Pelagius, as a resident 
in Rome, became acquainted with the works of this great Roman 
ecclesiastical figure. Some of the passages quoted are in all proba- 
bility the origin of the Pelagian passages like them, and it does not 
admit of the slightest doubt, I think, that the third passage, cited 
from the Aduersus Heluidium, is the origin of the corresponding 
part of Pelagius's commentary. The order in which they are given 
is the presumed chronological order of the works of Jerome. 

Epist. 18 B 4 (20) § 2 (a.d. 381 2): 'ne uideremur aliquid praeter- 
isse eorum quas ludaei uocant deuterosis^ et in quibus uniuersam 
scientiam ponunt.' Pel. in 1 Tim. i 4 fahulis. 'quas deuterosim 
appellant: unde in euangelio docentes doctrinas hominum condem- 
nantur. et genealogiis interminatis. generationibus antiquorum in 
quibus sibi summam scientiae uindicant ' 

Adu. Heluidium (about a.d. 383) Jerome first formulated the 
idea that 'brothers of the Lord' means 'cousins of the Lord.' Pel. 
on Gal. i 19 must therefore have got this view from Jerome^. 

^ This topic is also discussed in Quaest. app. 83 (pp. 476 f. of my 

2 I accept Vallarsi's dates for the letters without question. 

^ Reference to the articles devripuffis, deuterosis in the Greek and in the Latin 
Thesaurus respectively, will suggest other possibilities, but my view seems the most 

^ This was pointed out by Lightfoot, Galatians (ed. 186-5) p. 27*2, who acutely 
remarked 'unless his [i.e. Pel.'s) text has been tampered with here.' It has, but 
only to affirm the same view more definitely. 


Adu. Heluidium 6 (t. ii pp. 210 E, 211 A ed. Vall.^): 'nunc 
illud est ostendendum ut, quo modo ibi consuetudinem scripturae 
secutus est, sic etiam in doneo eiusdem scripturae auctoritate 
frangatur, quae saepe certum tempus, ut ipse disseruit, in eius 
adsumptione significat, saepe infinitum, ut est illud quod deus ad 
quosdam loquitur in propheta (Esai. xxxxvi 4^): "ego sum, ego sum, 
et donee senescatis ego sum." numquid post quam illi senuerint, 
deus esse desistet?' Pel. in 1 Cor. xv 25: 'donec autem non semper 
finem significat, sicut est illud: "ego deus uester donec senescatis," 
et cetera talia.' It ought to be mentioned that in Jerome, shortly- 
after the above passage, this very portion of 1 Cor. xv is quoted in 
its context; so that there is not the slightest doubt that Pelagius 
was indebted to Jerome's Aduersus Heluidium. By 'cetera talia' 
Pel. is referring to Ps. cxxii 2 (3) which Jerome afterwards quotes 
as a further illustration of this use of donec. 

Comm. in episL Gal. (about a.d. 386). Mangenot says: 'II est 
a noter que Pelage a connu le commentaire de saint Jerome sur 
I'Epitre aux Galates^.' 

De uiris inlustr. c. 5 (a.d. 392): 'epistula autem quae fertur ad 
Hebraeos, non eius creditur, propter stili sermonisque dissonantiara, 
sed uel Barnabae iuxta TertuUianum, uel Lucae euangelistae 
iuxta quosdam, uel dementis Romanae postea ecclesiae episcopi'', 
quem aiunt sententias Pauli proprio ordinasse et omasse sermone. 
uel certe quia Paulus scribebat ad Hebraeos et propter inuidiam 
sui apud eos nominis, titulum in principio salutationis amputa- 
uerat (scripserat autem ut Hebraeus Hebraice, id est suo eloquio 
disertissime), ea quae eloquenter scripta fuerant in Hebraeo, 
eloquer^tius uertisse in Graecura, et banc esse caiisam quod a 
ceteris Pauli epistulis discrepare uideatur^' Pel. argum. omn. epist.: 
'epistulam sane quae ad Hebraeos scribitur, quidam Pauli non esse 
adfirmant eo quod non sit eius nomine titulata, et propter sermonis 
stilique distantiam, sed aut Barnabae iuxta TertuUianum aut Lucae 
iuxta quosdam, uel certe dementis discipuli apostolorum et episcopi 

1 Vail. ed. 2 wrongly 43, but Victorias before him rightly 46. 
- Revue du clergi frangais, 1916, 1"' Avi'il et 1" Mai, p. 28 n. 1 of tirage a part. 
3 Cf. Hier. epist. 129 §3. 

* With the tenor of this passage of Jerome cf. also Filast. //aer. 61(89), and Isid. 
I'.tymol. VI 2 §45 (from Hier.). 


Romanae ecclesiae post apostolos ordinati,...ipsius magis esse cre- 
denda est quae tanto doctrinae suae fulget odium 
nominis fronte praelati. . .non est sane mirum si eloquentior uideatur 
in proprio, id est Hebraeo, quam in peregrine, id est Graeco, quo 
ceterae epistulae sunt scriptae sermone.' No case of borrowing 
could be clearer, and no mediaeval scholar could be blamed for 
attributing this 'argument' to Jerome. 

Prologus Galeatus (about a.d. 392) (ed. Vall.^ t. ix pp. 453, 454) 
'Syrorum...etChaldaeorum lingua... quae Hebraeae magna ex parte 
confinis est.' Cf. Pel. in 1 Cor. xvi 22 'magis Syrum est quam 
Hebraeum, tamen etsi ex confinio utrarumque linguarum aliquid 
Hebraeum sonat.' 

Adu. louin. I 26 (after a.d. 392): for a possible borrowing from 
this work, see on 1 Cor. viiii 5, under the section dealing with 
Ambrosiaster parallels. 

Epist. 52, 7 § 2 (a.d. 394): 'amare filiorum, tiraere seruorum 
est.' Pel. in Rom. viii 15: 'tiraere seruorum est, diligere filiorum.' 

Epist. 74, 3 § 3 (a.d. 398): 'Mattheus quoque in KuraXoyM 
apostolorum publicanus dicitur, non quo permanserit publicanus 
post apostolicam dignitatem, sed qui prius fuerit publicanus.' Cf. 
Pel. in Col. iiii 14: (Lucas) 'ex-medico erat, sicut Matheus iam 
apostolus adhuc dicitur publicanus.' 


At first sight the name of Augustine among the sources of 
Pelagius will seem even more incredible than that of Jerome, but 
we must remember that Augustine had been for some time before 
the close of the fourth century a writer of note, that he and Pelagius 
corresponded, and that it was only as the result of Pelagius's de- 
clared views that Augustine's developed in the course of controversy . 
Dr Smith's researches have shown that Ambrosiaster influenced 
Augustine as well as Pelagius, also that in Pelagius we find clear 
traces of the use of Augustine's Expositio Quarundam Pi'oposi- 
tionum ex Epistula ad Romanos and Epistulae ad Romanos Tnchoata 
Expositio, both written about A.D. 394. The more striking of these 
we proceed to give. 

Rom. i 18 — 19. Pel. quotes from Sap. xiii 5, while Aug. quotes 
from Sap. xiii 9. 


Rom. i 24. Pel. quotes from Ps, Ixxx 12: 'sicut in psalmo dicit: 
"et diraisi eos secundum desideria cordis eorum.'" Aug. writes: 
'quod autem dicit "tradidit," intellegitur "dimisit in desideria 
cordis eorum.'" 

Rom. V 14. Pel.: 'ut quidam dicunt: "forma" a contrario: hoc 
est: sicut ille peccati caput, ita et iste iustitiae.' Aug.: '"forma" 
autem "futuri" dictus est Adam, sed a contrario, ut quo modo per 
ilium mors, sic per dominum nostrum uita.' Aug. is clearly referred 
to here in 'quidam.' 

Rom. vii 2. Pel. says that by 'uir' Paul means the 'legis 
mandatum,' by 'mulier' the 'plebs' or the 'anima.' Aug. says 
that the 'mulier' corresponds to the 'anima,' while the 'uir' 
corresponds to the 'passiones peccatorum.' 

Rom. viii 26 f Pel.: '"postulat," quia postulare nos facit ge- 
mitibus qui enarrari non possunt, sicut temptare nos dicitur deus, 
ut sciat, hoc est, ut scire nos faciat, quales simus.' Aug.: 'gemere 
dicit spiritum, quod nos gemere faciat caritate, concitans desiderium 
futurae uitae, sicut dicit: "temptat uos dominus deus uester, ut 
sciat si diligitis eum," id est, ut scire uos faciat.' 

Rom. viiii 15 — 21. Pel.: 'ita non uolentis neque currentis 
tantum, sed et domini adiuuantis.' Aug.: 'non sufficere dicit uelle 
nostrum, nisi adiuuet deus.' 

Rom. xiiii 16, 22. Pel.: 'libertas quam habemus in domino, ut 
omnia nobis munda sint' (cf v. 20, also Tit. i 15). Aug.: 'bona est 
haec fides qua credimus omnia munda mundis' (Tit. i 15) (cf. also 
Aug. on Rom. xv 8, 9). 

The following refer to the Inchoata Expositio. 

Rom. i 4. Pel.: 'non omnium resurgentium, sed ad Christum 
pertinentium, in ipso Christo resurrectionis forma portenditur.' 
Aug.: 'neque (praecesserat) ad exemplum omnium resurgentium... 
qui praedestinatus est filius dei ex resurrectione mortuorum suorum, 
hoc est, ad se pertinentium, in uitam aeternam.' 

Rom. ii 2. Pel.: 'humanum iudicium multis modis corrurapitur; 
amore, odio, timore, auaritia saepe iudicii integritas uiolatur.' Aug. : 
'nam et indices mali praebent gratiam in accipiendis personis, 
aliqua cupiditate inlecti aut timore perterriti.' 

Rom. vi 9. Pel.: 'iam non potestis iterum baptizari, quia 
Christus non potest pro nobis iterum crucifigi, sicut dicit ad 


Hebraeos: "inpossibile est eos qui semel inluminati sunt," et cetera, 
quibus non paenitentiam negat, sed iterationem baptism! diffitetur/ 
Aug. : '. . .non posse deinceps eum qui peccauerit, iterum baptizando 
purgari; quo intellectu non intercluditur paenitendi locus... non 
enira possunt denuo baptizari qui semel baptizati sunt.' 

It did not fall within the scope of Dr Smith's investigation to 
examine further works of Augustine. I must therefore myself call 
attention to another work of Augustine, namely De Diuei'sis Quaes- 
tionibus ad Simplicianum (assigned to the year 397), as certainly 
used by Pelagius^ This fact the following parallels will make 

Book i, quaestio 1, deals with Rom. vii 7 — 25. 

Pel.: 'hinc in persona eius hominis loquitur qui legem accipit.' 
Aug. §1: 'quo loco uidetur mihi apostolus transfigurasse in se 
hominem sub lege positum, cuius uerbis ex persona sua loqui- 

Pel.: 'non dixit: "non habebam" aut "non faciebam," sed "nesci- 
ebam," hoc est, "nesciebam concupiscentiam " esse peccatum.' Aug. 
§ 2: 'itaque non ait: "peccatum non feci nisi per legem," sed 
"peccatum non cognoui nisi per legem." neque rursus ait: "nam 
concupiscentiam non habebam nisi lex diceret: non concupisces," 
sed ait: "concupiscentiam nesciebam nisi lex diceret: non concu- 
pisces." ' 

Pel,: 'quia iam sciens praeuaricaui.' Aug. §4: 'ut iama sciente 
et praeuaricante peccetur.' 

Book i, quaestio 2, deals with Rom. viiii 10 — 29. 

Pel. and Aug. (§3) both quote 2 Tim. iiii 7 'cursum consum- 
maui,' in this connexion. Pel. merely cites these words, while Aug. 
gives the whole of verses 7 and 8. 

Pel. and Aug. (§ 10) both quote 1 Cor. viiii 24 'sic currite ut 
[omnes] conprehendatis.' 

Pel.: 'non uolentis neque currentis tantum, sed et domini 

Aug. (§12): 'uoluntas hominis sola non sufficit ut iuste recteque 
uiuamus, nisi adiuuemur misericordia dei.' 

^ The reader should consult the independent article of the theologian Batiffol in 
the Revue Biblique nouv. s^r. t. xv (1918) pp. 5 ff. 

188 introduction [ch. 

Origen-Rufinus on Romans 

There is very little outward likeness between the Origen- 
Rufinus commentary on Romans (date about a.d. 405) and Pelagius. 
Origan as is well known, is discursive in this type of commentary 
and addicted to the allegorical method of interpretation. He com- 
ments on a group of verses at a time, while Pelagius deals with 
individual verses or clauses. Pelagius does not forswear the alle- 
gorical method, but he keeps it within bounds. His view about the 
matter is stated in his note on Gal. iiii 24 (cf 2 Cor. iii 6); 'ut 
manente historiae ueritate figuras testamenti ueteris exponamus... 
praecepta uero difficile inuenies apostolum taliter exponentem, ne 
eorum uideretur eneruare uirtutem.' He approves allegory in the 
same limited way as St Paul himself. Origen of course believed in 
historical interpretation also, but Pelagius in the proportion in 
which he allows the two methods, approximates much more to the 
Antiochian point of view. 

Yet there was a natural kinship of mind between Origen and 
Pelagius, as may be seen in their common attitude to Free Will. 
We should not be surprised to discover that Pelagius possessed 
enough Greek to tackle Origen 's difficult style. The view, however, 
that Pelagius consulted the original Greek of Origen's commentary 
on Romans, has been rendered exceedingly improbable by the in- 
vestigations of Dr Smith, who having proved use of Rufinus's 
'translation' of Origen on Pelagius's part, is justified in appealing 
to the old maxim 'entia non sunt multiplicanda praeter necessi- 
tatem,' and arguing that Pelagius used Rufinus only. Dr Smith 
has also studied the surviving fragments of Origen's Greek in 
Mr Ramsbotham's careful recension^, and has found no trace of 
their influence on Pelagius. It may therefore be taken as proved 
that it was Rufinus's adaptation- that was alone known to Pelagius. 
Incidentally this discovery enables us to be more precise with regard 

1 J.T.S. vols, xni (1911—12), xiv (1912—13). 

2 On Rufinus's methods of translation, consult E. J. Kimmel, De Rujino Ensebii 
interprete (Gerae, 1838) pp. 80 ff., Bp Westcott's article 'Origen' in D.C.B., Engel- 
brecht's edition of Rufinus's Gregory of Naziauzus (C.S.E.L. 46) pp. xviiiff., Ed. 
Schwartz's art. 'Eusebios' in Pauly-Wissowa, Real-Encycl. Bd. vi Sp. 1406, 
Mommsen in Eusebius-Rufinus Kirchengeschichte Bd. iii (Leipzig, 1909) pp. cclif., 
Koetzscbau in Origenes' Werke Bd. v (De Priricipiis) (Leipzig, 1913) pp. cxxviiiff., 
G. Bardy in Revue Bihliqne nouv. ser. t. xvi (1919) pp. 106 ff., t. xxix (1820) pp. 229 ff. 


to the date of Pelagius's commentary, if we may assume that he wrote 
the Romans portion first, an assumption which is a priori probable. 
The Pelagian commentary, then, belongs to the period 404 — 409 

Rom. i 1. Both commentators quote as examples of changed 
name, Abraham, Sara and Cephas, and both quote Phil, ii 7 in 
commenting on seruus lesu Christi. 

Rom, i 8. Both commentators refer to the expression 'God of 
Abraham,' and both suggest that the conversion of the Romans 
was matter for wonder. 

Rom. ii 1, 2. Pel: 'omnes norunt et innocentiam mereri prae- 
mium et malitiam habere supplicium.' O.-R.: 'communis professio 
est bonum non debere puniri nee malum consequi bona.' 

Rom. ii 28, 29. Pelagius shows here undoubted acquaintance 
with the very long note of Origen-Rufinus, for example, with the 
symbolical view of circumcision, and the parallel between Joshua 
and Christ. 

Rom. iii furnishes few parallels between the two commentaries. 
In iii 5 both contain references to the Deluge. At iii 13 Pel. has: 
'aliud ore promentes, aliud corde uoluentes'; O.-R.: 'dolus est 
cum aliud quis lingua loquitur et aliud uolutat in corde.' At iii 15 
Pel.: 'interficientes animas adulando': O.-R.: 'multo etiam uerius 
ille homicida dicendus est qui animam a uera uita separat.' 

Rom. iiii 8. Pel.: 'quidam dicunt remitti per baptismum, tegi 
laboribus paenitentiae, non imputari per martyrium.' O.-R. in 
Rom. ii 2: 'sed requiritur si erga eos quorum remissae sunt ini- 
quitates per baptismi gratiam, uel quorum tecta sunt per paeni- 
tentiam peccata, uel quibus imputandum non est peccatum per 
martyrii gloriam, secundum ueritatem uideatur deus seruare in- 
dicium.' This instance alone, in the absence of any other evidence, 
would suffice to show use of Origen-Rufinus by Pelagius. 

Rom. iiii 18. The question about Keturah which Pel. asks and 
answers here, was doubtless suggested by O.-R. on iiii 18 — 22. 

Rom. V 8, 9. Pel.'s 'commendat, amabilem facit' comes from 
O.-R.'s 'commendat uel confirmat intellegitur uel amabilem facit 
pro beneficiis praestitis.' 

Rom. V 12 — 14. Pel. points out that in saying 'omnes' Paul is 
exaggerating: O.-R. says that Paul makes it abundantly clear 
'omnes homines et multos homines idem esse.' 


Rom. vi 5 — 7. Pel: 'hoc est, ut omnia uitia destruantur, quia 
umim uitium merabrum est peccati, omnia corpus.' O.-R. : 'possunt 
autera membra ex quibus corpus istud peccati constat, ilia uideri 
quae superius enumerauit apostolus,' followed by an enumeration 
of twelve sins. In his alternative explanation Pel. defines corpus 
peccati as 'corpus nostrum.' O.-R. says: 'si uero magis hoc corpus 
nostrum dixisse intellegatur.' 

Rom. vi 7 — 9. Pel.: 'iam non potestis iterum baptizari, quia 
Christus non potest... iterum crucifigi, sicut dicit ad Hebraeos: "in- 
possibile est eos qui semel inluminati sunt," et cetera.' O.-R.: 
'" nos. . .qui semel inluminati sumus ..." (non) exspectemus ut " iterum 
post lapsum renouemur ad paenitentiam, rursum crucifigentes in 
nobis ipsis filium dei^'" Both also quote Col. iii 3. Pel.: 'mortuus 
enim omnino non peccat': O.-R.: 'qui enim...existimat...mortuum 
se esse, non peccat.' 

Rom. vi 13. Pel.: 'ut oculus qui ante uidebat ad concupiscen- 
dum, nunc uideat nudum ad uestiendum. sic etiam de reliquis 
membris aduerte.' O.-R.: ' circumspiciebant prius oculi mulierem 
aut alienum aliquid ad concupiscendum; nunc circumspiciant pau- 
peres debiles egenos ad miserandum.' 

Rom. vi 23. Pel.: 'non dixit similiter "stipendia iustitiae," quia 
etc' O.-R. t. VI c. 1 p. 239 L.: 'unde idem apostolus in alio 
loco: "stipendia," inquit, "peccati mors," et non addidit ut similiter 
diceret: "stipendia autem iustitiae uita aeterna," etc' 

Rom. vii 1 — 6. Pelagius's notes here appear to be based on the 
lengthy discussion of Origen-Rufinus. 

Rom. vii 7. Pel.: 'non dixit: "non habebam," aut "non faciebam," 
sed "nesciebam,"hoc est,nesciebam concupiscentiam esse peccatum.' 
O.-R.: 'non dixit: "non habebam," sed ait: ''nesciebam," tamquam 
quae esset quidem, ignoraretur tamen quod esset concupiscentia.' 

Rom. vii 16. Pel.: 'cum lege sentio, quae mala et non uult et 
prohibet.' O.-R.: 'consentit legi dei quia bona est, quae prohibet 

Rom. vii 18. Pel.: 'est uoluntas, sed non est effectus.' O.-R.: 
'est talis quaedam infirmitas (sc uoluntatis)...ut non statim uolun- 
tatem sequatur effectus.' 

Rom. vii 19. Pel: 'sicut, uerbi gratia, siquis iam diu iurare 
1 See also Aug. above, pp. 186 f. 


consueuit, etiam cum non optat, incurrit.' O.-R.: 'nee talis est 
quae dicat: "est, est: non, non," et ideo non potest operari quae 
uult, sed quae non uult,' 

Rom. viii 1. Pel: 'nihil in illis damnatione dignum est.' 
O.-R.: 'pronuntiat (sc. Paulus) in his nihil damnatione esse 

Rom. viii 3. Pel.: 'similem ergo ceteris hominibus carnem 
accepit, quantum ad naturam.' O.-R.: 'naturam quidem corporis 
nostri habuit.' On "et de peccato damnauit peccatum in carne" 
Pel. says: 'sicut hostiae quas pro peccato offerebant in lege, 
peccati nomine uocabantur...sic et Christi caro quae pro peccatis 
nostris oblata est, peccati nomen accepit'; O.-R. has '"et de peccato," 
uel — ut uerius habetur apud Graecos — , "et pro peccato damnauit 
peccatum in carne.'" As Dr Smith points out\ this is quite obviously 
an interpolation by Rufinus. 

Rom. viii 9. Pel. and O.-R. both quote the list of fruits of the 
Spirit from Gal. v 22, 23. 

Rom. viii 15. Pel. and O.-R. both quote Mai. i 6. Also, Pel: 
*qui uocat patrem, filium se esse profitetur': O.-R.: 'neque enim 
patrem alius quis nisi filius uocat.' 

Rom. viii 19 — 22. Pel. interprets 'jjreatura' here in the sense 
of 'angeli'; and this is one of the views that O.-R. mentions. O.-R. 
and Pel. also both use the expression 'rationabilis creatura.' 

Rom. viii 26 f. Both Pel. and O.-R. quote 1 Cor. xiii 12, and 
both also quote in illustration 1 Cor. xiiii 14 and 12: Pel: 'hie 
gratiam spiritus spiritum nominauit'; O.-R.: 'dona uel gratias 
sancti spiritus multos spiritus nominari'...' spiritum suum dicens 
gratiam sancti spiritus.' 

Rom. viii 28 ff. Here and elsewhere there is much common to 
the thought of Pel. and O.-R., even though resemblances in language 
are not easily found. Pel. was in some sense the inheritor and 
developer of Origen's position on various questions. 

Rom. viii 31 flf. Both Pel. and O.-R. quote 1 loh. iiii 18. 

Rom. viiii 14 — 19. Dr Smith indicates^ that Pel. and O.-R. 
are here agreed on three points: (a) vv. 14 — 19 are put into the 
mouth of an opponent of St Paul : (b) both insist on the freedom 

1 J.T.S. vol. XX (1918—19) p. 156. 

2 J.T.S. vol. XX (1918—19) pp. 163 f. 


of the will : (c) God's foreknowledge is foreknowledge of character. 
For a detailed comparison of the comments the reader is referred 
to Dr Smith's exposition. See also under Rom. x 15. 

E-om. X 4 — 11. Pel.'s comment implies the view of O.-R. that 
'God's righteousness' (v. 3) is the same as 'the righteousness which 
is of faith' (v. 6), while 'their own righteousness' (v. 3) is the same 
as 'the righteousness which is of the Law' (v. 5). O.-R. is probably 
one of the 'quidam' referred to later in Pel.'s notes. 

Rom. X 15. O.-R. on this verse: 'isti sunt pedes quibus et 
Paulus "cucurrisse cursum se" dixit, et "sic currere ut compre- 
hendat,'" recalls Pel, on Rom. viiii 16: 'si "non est uolentis neque 
currentis,"...quare et ipse cucurrit dicens: "cursum consummaui," 
et alios ut currerent adhortatus est dicens: "sic currite ut omnes 
conprehendatis " ? ' 

Rom, xi 7 — 10. On v. 8 Pel. writes: 'scriptura dicit: "ante 
hominem uita et mors; quod placuerit ei, dabitur illi," ne libertas 
scilicet tollatur arbitrii,' 0,-R. on Rom, i 24, 25, to which there is 
a cross reference from his note here, writes: 'seruatur ei (sc. animae) 
in omnibus libertas sui arbitrii, ut,..quodcumque uoluerit ipsa 
declinet, sicut scriptum est: "ecce posui ante faciem tuam uitam et 
mortem, ignem et aquam.'" 

Rom. xi 17. Both Pel, and O.-R, use the striking phrase 
'pinguedinis Christi,' 

Rom, xi 25. Pel.: 'occasionem eis salutis etiam "caecitas" prae- 
stitit "Istrahel."' 0,-R.: 'occasio, , ,conferendae in uos misericordiae 
populi "Istrahel" incredulitas exstitit,' 

Rom, xi 28, Pel,: 'inimici mihi sunt, quia nobis praedico 
Christum, sicut ipse alibi ait: "prohibentes nos gentibus loqui ut 
saluaefiant,"' 0,-R,: 'sed quod dicit: "propter uos," hoc est, quorum 
saluti scilicet inuident, "prohibentes" apostolos "gentibus loqui" et 
persequentes eos qui adnuntiant Christum,' 

Rom. xii 3. Pel. and O.-R. both quote 1 Cor. xii 11 in illustra- 

Rom. xii 6, Pel.: 'omnibus quidem credentibus gloria pro- 
mittitur in futuro, sed qui ita mundum cor habuerit ut hoc mere- 
atur, gratiam uirtutum accipit etiam in praesenti, quam deus ei 
donare uoluerit.' O.-R.: 'si enim in praesenti saeculo dat deus uni 
cuique gratiam "secundum mensuram fidei," sine dubio et in futuro 
dabit uni cuique gratiam pro mensura meritorum.' 


Rom. xii 8. Pel.: '"qui praeest" ecclesiae uel fratribus, debet 
esse sollicitus.' O.-R.: '"qui" uero "praeest" fratribus, uel "qui 
praeest" ecclesiae, "in sollicitudine " esse debet.' Again Pel.: 'tris- 
tem (sc. datorem) sine dubio odit.' O.-R.: 'non uult in tali opere 
esse tristitiam.' 

Rom. xii 17. Pel.: 'quod si tantae patientiae. ..fueris, non solum 
apud dominum, sed et apud omnes homines poteris probabilis 
apparere.' O.-R. on v. 19: 'patientia probabilis fit apud deum.' 

Rom. xiii 3. Pel.: 'ipsa damnatio malorum laus est bonorum.' 
O.-R. : 'potestas omnis adeo data est "ad uindictam quidem malorum, 
laudem uero bonorum.'" 

Rom. XV 1 — 3. Pel.: 'si uere firmi estis, sic facite ut ego, qui 
"factus sum infirmis infirmus, ut infirmos lucri facerem."' O.-R.: 
'uidetur in his Paulus firmum semet ipsum pronuntiare, sicut et 
in prima ad Corinthios dicit quia "factus sum infirmis infirmus, ut 
infirmos lucrarer.'" Pel.: 'non a nobis ipsis, sed a proximis con- 
laudemur.' O.-R.: 'sed non ex hoc accipias eum quasi inmemorem 
mandati illius loci quo dicitur: "laudet te proximus tuus et non 
tuum OS, extraneus et non tua labia." (Prou. xxvii 2.)' Pel.: 'sicut 
et alibi suum nobis proponit exemplum, dicens: "sicut et ego om- 
nibus per omnia placeo, non quaerens quod mihi utile est, sed quod 
multis, ut salui fiant."' O.-R.: '...quod in aliis dicit: "sicut et ego 
omnibus per omnia placeo, non quaerens quod mihi utile est, sed 
quod multis, ut salui fiant (1 Cor. x 33).'" 

Rom. XV 5 ff. Both commentators quote Matth. xviii 19. 

Rom. XV 17. Both commentators quote 1 Cor. i 31. 

Rom. XV 24. Pel.: 'ideo ex parte, quia nulla magnitudo tem- 
poris satiat caritatem.' O.-R.: 'quam tamen caritatem tantam 
praesentit futuram cui nee possit ex integro satis fieri,' 

Rom. xvi 16. Both commentators refer to the kiss of Judas. 

Rom. xvi 20. Both commentators quote the Gospel passage 
about stepping on serpents and scorpions. 


Long ago Simon declared: 'il [Pelage] suit d'ordinaire les 
interpretations des Peres Grecs, principalement celles de Saint 
Chrysostdme',' and recently Zahn has repeated the statement: 

1 Histoire critique des...Commentateurs, p. 242. 
S. P. 13 


'Pel.. . .griech. Ausleger, im Gal, besonders unverkennbar den Chrys., 
beriicksichtigt hat^' There is no need to elaborate the proof of 
this: a few illustrations will suffice-. According to Zahn, Pelagius 
read in Rome about 404' the commentary of Chrysostom on Galatians, 
which was written before 3981 

Pelagius's identification of 'Predestination' with 'Foreknow- 
ledge' (Rom. viii 29) is traced by Simon to the Greeks •*, but this 
statement must now be qualified by Dr Smith's discovery of this 
doctrine in Ambrosiaster and Augustine^. 

Pel. in Rom. viiii 16: 'hie interrogantis uoce utitur et re- 
darguentis potius quam negantis': Chrys. (t. ix p. 614 b Bened.): 
iraXiv erepav avriOecnv eia-dytov, koX \ey(ov 'apa ovv ov rov OeXovro'i' 
/CTX.....etTa iraXiv avridecriv eTrdyet' 'dpa ovv ov BeXei iXeeV kt\.^ 

Pel. in Phil, iiii 15: 'dantes carnalia et spiritalia accipientes.' 
Chrys. (t. XI p. 313 C Bened.) (iKotvoovrjaav) et? Xoyov Soaeoif; rcov 
(xapKiKcov, Kul XT^i/retw? rdov TrvevfiariKcov. 

Pel. in 2 Thess. i 10: 'quia multi de die illo nostro testimonio 
credidistis.' Chrys. (t, XI p, 522 a): /jiiWcov Se inrep irdarj^; t^? 
f&)j}9 aov Xoyuv Bi^ovai, kol €v6vva<i vire'^^etv, oOSe aXXoyv <re 
dvap,i fjLvrjaKovTcov virep T-ij^; 8i,k7}<; dve')(^r] ; kt\. Both thus refer this 
clause to the Day of Judgment. See Swete on Theod.-Mops, ad loc. 
Pel. and Chrys, seem to be the only two ancients who regard in die 
illo as part of the preceding clause. 

Pel, in 2 Thess. ii 3: 'nisi antichristus uenerit, non ueniet 
Christus. quod autem "discessio" hie dicit, alibi eum "refugam" 
appellauit in Latinis exemplaribus: utrumque autem ita intelle- 
gendum est quod "nisi uenerit refuga" ueritatis, sine sui principatus 
desertor, sine discessio gentium a regno Romano, sicut in Danihelo 
per bestiae imaginem dicit.' Chrys. (t. XI p. 525 b): irepl tov 
dvTtXP^o-TOv ivravda hiaKeyeraL.Kai p-eydXaaTTOKaXvirreL iJivaTt^pia. 
Tt iariv 'rj diroa-Tacria ; avTov KoXel diroaraaiav, w? ttoXXov^ 

1 Zabn, Der Brief d. Paulus a. die Galater 2 Aufl. (Leipz. 1907) p. 25. 
- Incidentally a considerable number of agreements between Pel. and Chrys. will 
be found stated in the notes to Swete's Theodore of Mopsuestia. 
3 Zahn, ibid. * Simon, p. 242. 

5 J.T.S. vol, xrx pp. 201 f. 

6 Simon, p. 242, wlio here however cites the interpolator of Pel., who is much 
fuller than Pel. himself. Elsewhere also Simon's arguments are invalidated some- 
what because he could not distinguish interpolations from the original Pelagius. 


/xeWopTU aTToXkvvat kuI dcpcardv. Later Chrys. denies the identity 
of Antichrist and Satan. Theod.-Mops. and others follow Chrys. 

Pel. in 1 Tim. iii 11 started in the West an interpretation which 
is found in Chrys. etc. The quotations are: 'similiter eas ut dia- 
conos eligi iubet: unde intellegitur quod de his dicat quas adhuc 
hodie in oriente diaconissas appellant.' Chrys. (t. xi p. 605 a) : 
'<yvi>alKa<i oicravjw^! Bia/covovi <pr]CTiv...'7r€pl twv to d^iwfia tt}? 
Bi,aKovia<i i'^ovacou (^rjcriv. So Theod.-Mops. 

Many more parallels between Pelagius and Chrysostom might 
be adduced, as is done by Swete in his notes to Theodore of 
Mopsuestia, but I have tried as far as possible to avoid mentioning 
cases where Theodore and others support Chrysostom's view. For 
our purpose it is necessary especially to collect instances that lack 
further Greek support. The fact that Pelagius was really influenced 
by Greek authors is nowhere more conspicuously evident than in 
his treatment of the difficult phrase in Phil, ii 6 non rapinam 
arbitratus est esse se aequalem deo. A reference to Lightfoot's well 
known catena of patristic opinion on this clause shows that Pelagius 
with his 'quod erat humilitate celauit, dans nobis exemplum ne in 
his gloriemur quae forsitan non habemus' belongs to the Greeks, 
and not to the Latins. 

Theodore of Mopsuestia 

Theodore of Mopsuestia, who lived till 428, held Pelagian views, 
was personally acquainted with Pelagius, and wrote in Greek a 
commentary on all the Epistles of St Paul. We should expect to 
find parallels between such a commentary and the exposition of 
Pelagius. Unfortunately Theodore's commentary, as a connected 
work, has perished, except in a Latin translation of the ten shorter 
epistles, Galatians to Philemon \ The preservation of even this 
portion is due to the happy accident that some mediaeval scholar 
possessed an imperfect Ambrosiaster, and somehow managed to 
complete it by acquiring an anonymous copy of the Latin Theodore, 
which he caused to be copied as a continuation of Ambrosiaster on 
Romans, First and Second Corinthians. 

The fact that there are parallels between the Latin Theodore 

and the Pseudo-Jerome has naturally not escaped Professor Swete, 

1 Portions are preserved also in Isho'dad of Merv (Camb. 1916). 



whose edition of the Latin Theodore is a light on the path of the 
investigator of patristic exegesis. Our task is greatly simplified by 
his labours, and nothing need be done to show that there is a 
relationship between the two commentaries, save to make a critical 
revision of his Pseudo-Jerome text and to add from his notes some 
instances which he refrained from quoting in his introduction ^ 
It must, however, be remembered that Theodore was acquainted 
with the earlier work of Diodorus of Tarsus^ and of Chrysostom', 
a fact which complicates investigation. 

It is a more difficult question which of the two, if either, is the 
borrower. A priori, if borrowing there has been, it is more likely 
to have been on the side of the Latin Pelagius than of the Greek 
Theodore. But Theodore knew some Latin^, and in view of the 
intimacy of the two men the other possibility cannot be denied. 
We have no information as to the date of Theodore's commentary 
to help us to a decision. 

Gal. ii 2. Pel.: '"ne forte"... hoc non est dubitantis.' Theod.: 
'nam quod dicitur "ne quoquo modo" non dubitationis causa dicitur.' 

Gal. vi 11. Pel.: 'intellegite quam non timeam qui litteras mea 
manu perscripsi.' Theod.: 'designans quoniam neque ueretur eos, 
neque negat ilia sua esse quae dicit.' (The Greek also survives.) 

Eph. i 21. Pel.: 'quia unum est iam cum deo adsumptus homo.' 
Theod.: 'de suscepto homine id dicens, eo quod propter inhabitantem 
in eum naturam dei uerbi ab omnibus habet adorari.' 

Phil, i 2. Pel.: 'hie "episcopos" presbyteros intellegimus: non 
enim in una ciuitate plures episcopi esse potuissent.' Theod.: 
'"episcopos" dixit illos qui nunc presbyteri dicuntur; enim 
ordinis erat multos in una ciuitate esse illos qui nunc episcopi 

Phil, ii 5 ff . Pel.: 'multi praeterea hunc locum ita intellegunt 
quod secundum diuinitatem se humiliauerit Christus..."formam 
serui" hoc est, naturam hominis induendo..."semet ipsum exina- 
niuit": quod erat humilitate celauit.' Theod.: 'deus uerbum talia 
sponte fecisse uidebatur ita ut pro aliorum salute praehonorandara 
omnibus existimaret humilitatem . . . " formam " autem " serui," ut dicat 

^ Vol. I pp. Ixxiv — Ixxvi. - See Swete, vol. i pp. Ixxiii f. 

^ Swete, vol. i p. Ixxii ; Wohlenberg ed. of Pastoral Epistles (Leipzig, 1906) p. 76. 

* As his note on 2 Tim. iiii 13 (fie/x^odvas) shows. 


"naturam seriii," humanam sic uocans naturam.'...' (to ovv "kavrov 
eK6vo}(r€v," avrl tov ovk eSei^ev kavrov . . .rr]v d^lav eKeivrjv dire- 

Phil, ii 10. Pel.: 'ut omnes simul hominem cum uerbo adorent 
adsumptum.' Theod.: 'ut omnes ilium adorent... propter illam 
copulationem quam habet ad unigenitum.' 

Phil, iii 1. Pel: 'eadem repetere quae iam praesens dixeram.' 
Theod.: 'non quia iam scripserat dicit...sed quia docuerat eos 
instantissime... frequenter uobis locutus sum.' (The Greek exists.) 

1 Thess. ii 7. Pel.: 'id est, ut de euangelio uiueremus.' Theod.: 
'magis enim erat ut et habentes potestatem quasi apostoli ilia quae 
ad usus necesse habebant sibi acciperent' {fiel^ov yap rjv to kuI 
e')(ovra<i i^ovalav co? d7roar6Xov<i ra irpo'i ttjv ^peiav Xafx^dvetv). 

1 Thess. iiii 6. Pel.: 'in quolibet negotio nequis alterum fraudet.' 
Theod.: 'pudicissiine quidem dixit "in negotio'" etc. 

2 Thess. ii 1. Pel.: '"et nostrae congregationis in ipsum." 
quando a quattuor uentis caeli coagregabuntur electi, ut, ubi 
fuerit corpus, illic concurrant et aquilae.' Theod.: 'hoc ergo dicit: 
*'et nostram congregationem ad eum," dicit etenim Christiaduentum 
et congregationem nostram tunc ad eum frituram.' (on irepl Tr}<; 
l^picrrov 7rapov(Tla<; Kal tt}? rjfMoov eV avrov e'/cetV?;? e7ri(Tvi>ay(i}yrj<;.) 

Col. i 15. Pel.: '"primogenitus" secundum adsumpti hominis 
formam, non tempore, sed honore, iuxta illud: "filius meus primo- 
genitus Istrahel.'" Theod.: '"primogenitus" non tempore dicitur 
solum, sed et praehonoratione fr-equenter.' (The Greek exists.) 

1 Tim. i 15. Pel.: 'quern omnes credant et omnium conscientiae 
uerum esse cognoscant.' Theod.: 'omnis quicumque fuerit ille 
recipiet, credens quia deus homines misericordia saluat... omnis 
autem quicumque ille delectatur in hisce sermonibus....' 

1 Tim. ii 12. Pel.: 'publice non permittit: nam filium uel 
fratrem debet docere priuatim.' Theod.: 'neque uetabat mulieres 
ut impios maritos suos ad pietatem uel inuitarent uel docerent, 
aut pios inconuenienter conuersantes ad opera inuitarent uirtutum.' 

1 Tim. iii 1. Pel.: 'ad boni operis desiderium eum prouocat, 
non honoris.' Theod.: 'bene "opus" dixit et non dignitatem.' 

1 Tim. V 3. Pel.: 'necessaria praebendo uel solaciis fouendo.' 
Theod.: 'quod hoc in loco dixit: "honora," hoc est, diligentiam illis 
adhibe,' etc. 


1 Tim. V 9: Pel. confuses between the orders of widows and 
deaconesses, and Theod. refers to those that fall into this error. 
Pel.: 'tales uoluit eligi diaconissas, quae omnibus essent exempla 
uiuendi.' Theod.: 'quidam uero non considerantes quam ob causam 
aetatem uoluerit significari, hoc statuerunt utrumnam mulieres 
diaconissas ante hanc aetatem ordinari minime conueniat.' If we 
could be certain that Theod. was here criticizing Pelagius, the 
question of priority would be settled. Then Ave might argue that 
Pel. had come into Theod.'s hands at a time when he had got 
through all the epistles down to Colossians; for it is from 1 Thessa- 
lonians onwards that parallels between the two begin to be fairly 
numerous. Other passages where Pel. and Theod. disagree, are 
2 Thess. i 8—10; Tit. i 5—7, ii 14. 

1 Tim. V 24 — 25. Pel.: '" subseqiiitur eorura in {\itnvoh\dicium" .. . 
"quae aliter se habent"...etiam quae ad tempus latent, non possunt 
diutius occultari.' Theod.: 'sicut delinquentium hominum et non 
recte uiuentium delicta manifesta sunt, quae necessarie illis in 
futuro saeculo poenas sunt prouisura, licet si et faciant aliqua quae 
multos latere poterunt' etc. {wairep ra twv ovk ev ^lovvnov dv- 
dpcoTToyv TrraicTfiara TrpohrfKa eanv dvayKalw^; rrjv eVl tov fxeWovrof 
alcoi'o^ ri/xcopiav avroi'i eird<yovTa, el Koi (co9 etKo^;) nvd Trap" avrwv 
\av6dvei, rov<i ttoXXou? kt\.) 

1 Tim. vi 6. Pel.: '"cum sufficientia" non luxuriae, sed naturae.' 
Theod.: 'necessitatibus nostris sufficientes,' 

1 Tim. vi 20. Pel.: 'fidei "custodi depositum."' Theod.: 
'"custodi" fidem.' 

2 Tim. ii 6. Pel.: 'ostendit eum primitias fructuum a populo 
debere percipere, praeter mercedem quam a domino messis accipiet 
in futuro.' Theod.: 'quae ad usum tuum necessarie pertinent, 
indiscrete adsequere st fidelibus qui praestant tibi,' etc. 

2 Tim. ii 8. Pel.: 'hoc contra illos qui carnis resurrectionem 
negantes, consequenter etiam Christi negabant.' Theod.: 'ita ut 
nee resurrectio uera facta esse susciperetur; qui enim fieri poterat 
ut uera crederetur resurrectio, si caro uera ratione non fuisset 
sumpta?' etc. 

2 Tim. ii 15. Pel.: 'ille "recte tractat" qui dicta sua exemplo 
confirmat.' Theod.: 'recto edoce instituto, et non pigeas laborem 
pro his subire.' 


Tit. i 7. Pel.: 'ipsum dicit "episcopum" quern superius pres- 
byterum nominauit.' Theod.: 'nam duni dicit: "ut constituas... 
presbyteros," et de presbyteris disputans adiecit: "oportet eniiu 

Tit. i 15. Pel: '"coinquinati et infideles" sunt qui non credunt 
iugum legis a collo credentium dominum abstulisse.' Theod.: 
'siquis uero per suam incredulitatem habet conscientiam coin- 
quinatam, nihil mundum esse poterit illi qui talis est.' (Gk. also.) 

Tit. iii 15. Pel.: 'qui nos fideliter amant. Siue: Qui fidei causa 
nos diligunt.' Theod.: 'qui per fidei familiaritatem cum eo iunctas 
amicitias habere uidebantur.' 

Subsidiary Sources 

This paragraph is intended to include merely a few notes that 
have occurred to me in the course of repeated readings. The first 
and second concern the subject-matter, the remainder have to do 
with the expression, and reflect Pelagius's reading of classical lite- 

Pel. in Eph. ii 15: '"euacuans." per solam fidem iustiticans et 
moralia sola decernens.' Severianus, quoted by Swete, Theod.-Mops. 
ad loc: 'ev Soy/xaatv toi<; l8ioi<i [explaining these to be the prin- 
ciples of Christian morality].' 

Pel. in 2 Tim. ii 18: '"cancer" esse dicitur uulnus quod in 
mammillis nascitur feminarum, quibus nisi cito subuentum fuerit, 
cum uirus ad cor serpendo peruenerit, nullum ultra remedium est.' 
This is evidently a quotation or a recollection of some medical work\ 
cf Aug. ciu. dei xxii 8 (p. 570, 11, 14 ff. Domb.''): 'Innocentia, 
religiosissima femina, mamilla cancrum habebat, rem, sicut 
medici dicunt, nullis medicamentis sanabilem^.' 

The following appear to be undoubted recollections of classical 

Lucretius: i 149 — 150 'principium cuius hinc nobis exordia 
sumet, I nullam rem e nilo gigni diuinitus umquam' (cf, in 

1 Thes. Ling. Lat. s.v. cancer, vol. in p. 231, esp. 11. 47 ff., has a number of 
references, going as far back as Ov. met. ii 825, to the incurableness of this disease. 
- From which Isid. Etym. iv 8 § 14 comes. 


417seqq.): Pel. in Col. ii 8 'philosophos...dicentes ex nihilo fieri 
nihil posse': i 304 'tangere enim et tangi, nisi corpus, nulla potest 
res': Pel. in 1 Cor. vii 4 'corpus dicitur omne quod tangitur.' 
Knowledge of Lucretius is particularly interesting, for in the 
patristic period writers range between the exhaustive knowledge 
shown by Amobius and the absolute ignorance shown by the 
majority. Tertullian also (adu. Marc, iiii 8, De An. 5) quotes this 

Virgil: iiii 174, 188 'fama, malum qua non aliud uelocius 
ullum...tam ficti prauique tenax quam nuntia ueri': Pel. in 1 Thess. 
i 8 'natura famae haec est, ut sine bonum sine malum nuntians 
ubique omni celeritate discurrat.' iiii 569 — 570 'uarium et 
mutabile semper femina': Pel. in 1 Cor. xvi 13 'muliebris omnis 
inconstantia et uarietas iudicatur.' All Latin writers posterior to 
Virgil show more or less of his influence. 

Horace: epist. ii 1, 250 — 1 'sermones...repentis per humum': 
Pel. in 2 Cor. iiii 2 'uerbo dei sermonem uilem admiscent et per 
terram repentem^' 

Juvenal: 1, 142 'poena tamen praesens': Pel. in Gal. v 6 
'praesentem metuit poenam^' (Cf also Cic. and Plin., quoted by 
Mayor, and Sen. Phaedr. 162.) We know from Ammian that Juvenal 
was much read in the second half of the fourth century^ 

1 A reference to mulcere in lexx. will suggest possible imitation of other poetical 
passages by Pel. in I Cor. xiiii 7. 

- In his epist. ad Demetr. 25 (Migne, P.L. xxx 40 d) uacuus uiator et nudus non 
timet latronis insidias is an echo of luu. 10, 22 cantabit uacuus coram latrone uiator. 

^ Of. Duff's Juvenal, p. Ii. 




Certain of the existing authorities for the reconstruction of the 
text of Pelagius have already become partially known to the reader 
in the course of the discussion in the second chapter, but it is 
necessary to give here a fuller account of these, and at the same 
time to consider others. Further authorities, such as a number of 
late mediaeval compilations, based probably on some manuscripts 
of Pseudo- Jerome, it has not been possible to consider, for the simple 
reason that some limit must be set to these researches. I believe, 
however, that I have secured a broad and sound enough basis for 
the constitution of the text, in the authorities now to be described. 

The first place rightly belongs to the two (three) manuscripts 
of the original form, with the Vatican and Freiburg fragments. 
Then we shall take the St Gall MS 73, the Paris MS 653, the two 
families of Pseudo-Jerome MSS, and Cassiodorus. Finally must 
come the Wtirzburg and other glosses, Claudius of Turin, Zmaragdus 
of St Mihiel, Sedulius Scottus of Liege, Haymo of Auxerre, and 
later authorities. An attempt will be made to show the inter- 
relationship of various authorities and to estimate their relative 

(a) The Manuscripts of the Original Form 
(1) Codex Augiensis GXIX at Karlsruhe (A) 

The manuscript is thus described by the late Dr Alfred Holder 
in his great catalogue. Die Reichenauer Handschriften heschriehen 
und erldutert, P^ Band, Die Pergamenthandschriften (Leipzig, 1906) 
pp. 303 f , as modified in IP' Band (1914), pp. QQQ f. 

CXIX. 164 folia (2 columns, with 37, 36, 35, 33 lines to the 
page), measuring 302 by 221 mm., saec viil ex. — ix in. Five hands: 




the fifth scribe an old Irishman: written on the Continent; arche- 
t3^e Italian uncials^ of the fifth or sixth century; folia 149—164 
(1 column and 33 lines to the page), saec. x. 

f. 1 — 8 = 1 quaternion 1(a); 
9 10 11 X 12 13 X I 14 X 15 16 X x 17 B: 

18 — 83 = 2 quaternions CD; 

34 X 35 36 37 I 38 39 x 40 41 U ; 

i I i I ! ! ! i I 1 

42 — 49 = 1 quaternion; 50 — 57 = 1 quaternion VII; 
58 59 60 61 I X 62 63 64; 

65 66 X 67 68 I 69 70 71 x 72 UIIII ; 

73 74 75 76 I 77 78 x 79; 

I ! I I I 1 I 

80 — 87 = 1 quaternion; 88 — 95 = 1 quaternion XII; 
96 97 X 98 i 99 100 101 102; 

103 — 110 = 1 quaternion; 111 — 142 = 4 quaternions; 

143 144 145 146 I 147 148 x x; 

I I I I I I I i I 

149 — 164 = 2 quaternions. 

Pelagii commentarii in epistulas Pauli. Cf. Alexander Souter, 
The Commentary of Pelagius on the Epistles of Paul [from the 
Proceedings of the British Academy, vol. ii (Ld. 1907)] pp. 15 — 18, 
25—27. The Journal of Theological Studies, vol. viii (1907) 
pp. 535 — 536. 

1 I should now rather eay 'half-uncials': see below. 


f. 3, 1. Primum qugritur quare post euangeliflm: quae supple- 
mentuni legis sunt et in quibus nobis exempla et prgcepta uiuendi 
plenissime digesta sunt uoluerit apostolus has epistulas ad singu- 

las ecclesias destinare ut initio, nascentis ecclesig nouis causis 

(147' l)...excipit apostolus quos salutet; 

Explicit ad Tituin incipit ad Philemonem. 

Cui apostolus a Roma...debeamus. 

Paulus uinctus...(148, 2)...fidei societate. 

Fixplicit ad Philimonem. 

Expositio epistolae ad Hebraeos. 

149. Incipit argumentum ad Hebreos. 

Inprimis dicendum est cur apostolus... conposuit. 

Multifarig multisque modis olim dews loquens patribws in 

Ac si dicat per multos...(164')...paenitenti§ et perseuerentiae ; 

Finit expositio epistolae Pauli apostoli ad Hebreos. abot^e in 
nomine sanc^ae trinitatis semper nobis cum amen trina maigestas 
domni per quem fiat fiat. 

1'. ferur | ferunt | ferunt 'Ophyr conuexa. 

2. omnia mihi licent sed non omnia expediu(nt). 

3. liber monasterii Augie maioris. 

Title on the cover Expos?Yiones in eppi's^olas Pauli ad Roma??os, 
Corinthios, Galathas, Ephesios, Philippenses Tesallontcenses Colo- 
censes Thimothei<??i Titum Philomone??! Ebreos. 

Wooden cover, covered with grey leather; 2 leather clasps. 

It is of some importance to show how Holder in 1914 arrived 
at certain of the conclusions stated above. Holder's catalogue of 
the Reichenau manuscripts at Karlsruhe was itself in MS for a 
considerable period before it was printed, and I had been privileged 
to handle and consult it at Karlsruhe in that form in 1901, when 
engaged on the Ambrosiaster Quaestiones. Early in 1905 Dr C. H. 
Turner was at Karlsruhe pursuing his own researches, but found 
time to note from Holder's MS catalogue that there was in the 
Reichenau collection an anonymous manuscript of "the Pseudo- 
Primasius commentary." I have in my possession the first proof 
(1905) of the printed catalogue, in which MS cxix is described as 


such, I welcomed the information Dr Turner kindly gave me, 
and also the proof-sheet which my friend Holder sent me at my 
request. I saw immediately that the description of the MS was in- 
correct, and wrote to Dr Holder at once that his MS must be either 
a pure Pelagius or an anonymous Pseudo-Jerome. The loan of the 
MS, for which I asked, was refused by the Baden Minister of Public 
Instruction, and it was therefore necessary to consult it on the spot. 
A morning's work in July 1906 sufficed to show that a pure Pelagius 
had at last been discovered. I convinced Holder that his ascription 
was wrong, and in the first volume of the catalogue, as published 
in September 1906, the manuscript is described as: (s. ix): 
Pelagii (Walahfridi Strabi) commentarii in epistulas Pauli. 

During the interval between that date and 1914 he advanced 
yet further, put the date back from saec. ix to saec. viii ex. — ix in., 
accepted my statements that five scribes wrote the MS on the 
Continent, of whom one was an Irishman, and that the archetype 
was an Italian uncial of the fifth or sixth century, and struck out 
'Walahfridi Strabi' altogether. 

These are the facts, and it has been necessary to state them 
exactly, because from the way in which the Abbe Mangenot writes, 
the reader would almost certainly infer that Holder had first 
described the manuscript in the published catalogue as Pelagius, 
and that then, and not till then, had I come to know of the 
manuscript as such: whereas my collation of it was complete at 
least two and a half weeks before the catalogue Avas published ^ 
An apology is perhaps necessary for insisting so strongly on a 
personal claim. 

From the collotype of one page wdiich was published in the 
seventh volume of the Proceedings of the British Academy^, the 
reader will see that the script in which it is written is the ordinary 
pre-Caroline minuscule of the district Murbach-Reichenau-St 
Gallen-Chur, to which Traube has called attention*. It need not 

1 'Holder signala un nouveau manuscrit de V Expositio de Pelage, du tx« si^cle... 
Alexandre Souter entreprit une edition critique' (Mangenot in Revue du Clerge 
fratigais for 1916, p. 20 of tirage a part). The collation was completed ou Aug. 14, 
1906; the catalogue appeared in Sept. 1906. 

- The frontispiece of my second paper in the Proceedings (p. 1 = 261). 

■^ Abhandlungen d. Kgl. Bayer. Akad. der Wiss., philos.-philol. u. histor. Kl. 
Bd. XXV (2) (1910) p. 52. 




be doubted that the MS was written at Reichenau itself about the 
end of the eighth or the beginning of the ninth century. 
The scribes partition the work between them thus: 

(1) K 3 — 83 (gentium plenitudo id). 

(2) ff. 34 — 59 b (pater Jilius ad ettate). 

(3) ff. 59 va — 100 a (et ideo ipse paries ini-). 

(4) ff. 100 b— 106 a 1. 4 (uere est in uohis). 

(5) ff. 106 a 1. 4—148 b. 

(3) is the most beautiful of all, (4) was possibly Irish, and (5) 
was, as has been said, Irish : the forms of the capitals P (e.g. f, 126 a), 
B (e.g. f. 133 va) and >S^, for example, are Irish, and there is an Irish 
r on fol. 124 a. 

The scribes make very little use of abbreviation or contraction. 
The following list^ must be approximately complete: 

autem au (first and third and fifth 
scribes), aut (second and fifth 
scribes, corrector of fifth scribe) 
Chrishis xps 
deus ds 
dieit die 
dixit dix 
dominus dom (second scribe, six 

times), dns 
eius es^ (ff. Ill vb, 114b) 
esse ee 
est e 

fratres ffs (third scribe) 
lesus ihT 

Istrahel isral (second scribe, twice), 
isrl (third scribe), irl (fifth scribe, 
f. 126 vb) 
mens ms (fifth scribe, end of line) 
uohis uob (third scribe) 
non n 

nosier ^■ = nostrum (f. 40 va) •N = 
nostri (f. 101 va) 
nt (third scribe, f. 65 b, 73 va, 
74 vb, 76 a, 92 vb, also fifth 
scribe): so ut ( = 71 b) 
ni, nm 

nfm (rare),nfae,nfas : so ufm, 

nrum : so urum (fifth scribe 

'^o&{l) = nostram (f. 114 vb) 
not = noster (f 119 va)^ 
omnes om 
per p 

post p', p' (third scribe) 
prae p (third scribe only) 
pro <p 
propter ppt (second and third scribes), 

pp (third scribe, f. 70 a only) 
quae q:- qu; qy 
que q; q: <^ (second scribe only, 

f. 50 va, vb) 
q^d q (third scribe only) 
quod qd (end of line three times, 

first scribe), qod (second scribe) 
quoniam qm (first scribe only, rarely), 

qnm (first and third scribes only), 

quo (second, third and fifth scribes) 
saeculuni sclm 
sanctus scs 
sicut sic (first scribe) 
spiritus sps 
sunt H 

uel ul (first scribe) 
uero uo (third scribe, end of line) 

^ I should mentiou here once for all that the lists of abbreviations given in this 
book were put at the disposal of Prof. W. M. Lindsay for his Notae Latinae (Cam- 
bridge, 1915). 

2 The only instances known to Lindsay (Notae Latinae, p. xv); it is on the 
analogy of hs {huius) cs [mms). -^ See Traube, Nomina Sacra, p. 234. 


Syllable Symbols: 

con c (third and fifth scribes only) xir t' (second, third and fifth scribes), 

en m 'men' (first scribe only) (t2 third scribe) (nar 'mur,' end of 

er t 'ter,' u 'uer' (third scribe only) line) 

is b 'bis,' 1 'lis' (third scribe only) yg b; m; 

it c X (first scribe) ' (third scribe) 

m suprascript stroke rt^ (third scribe) 

rum T% (second and third scribes), l=lus (end of line, third scribe) 

if (fifth scribe) \ (f. 79 va) (third scribe) 
runt f (first scribe) 

Other palaeographical features of interest are these. The long 
i is never used. The ligature ti is used irrespective of the distinc- 
tion between assibilated and unassibilated ti\ thus in injiati and 
generatis, as well as in gratiae. The following combinations of 1 etters 
are sometimes in ligature, the first and second very often: n", nt, 
at, an: the second scribe puts us in ligature once at the end of the 
line: the ^ for ae is quite often found. All the scribes employ a big 
Z, which lies on the line, with only the tail below. 

Certain of these abbreviations shed light on the character of 
the MS from which the Reichenau MS was copied. 

dorii = dominus (ff. 36 va, vb, 87 va, 38 b, 40 va, 47 vb, 79 b) 
comes straight from the archetype, being almost obsolete at the end 
of the eighth, and the beginning of the ninth century. It is in fact 
a specialty of the fourth, fifth, sixth and seventh centuries^ 

u = any case of noster (flf. 40 va, 101 va, 141 va) was rarely used 
after the middle of the sixth century, and never naturally after the 

This evidence is borne out by instances of the preservation of 
really old orthography which had been given up long before the 
date of our MS. 

The long forms idololatria etc. are rarely found in MSS of this 
date, but are almost invariable^ in our codex (e.g. ff. 6 vb, 8 b, va, 
9 a, 63 a). The solitary* MS of Lucifer of Cagliari (cod. Vatic. 133, 
saec. IX — x), the Laudian MS misc. 130 (saec. ix — x) of Augustine 
De Baptismo, also furnish it. There can be little doubt, I think, 

^ See Traube, Nomina Sacra (Miinchen, 1907) pp. 168 ff. 

2 The latest examples known to me (apart from those in Traube op. clt. p. 207, 
and one or two noted below) are Koln MS ccxii (saec. viiex.) (Canons) in Chroust's 
Monumenta Palaeographica no. 298 where N occurs for noster and nostrum, and 
Milan, Ambros. O 210 Sup. (saec. vi ex.) (Acta Archelai, ed. Beeson (Lpz. 1906) p. xx). 

' An exception, f. 96 a. ^ But see now Rev. Bened. xxxiii (1921) pp. 121 ff. 


that the syncopated form is not really ancient. It would be in- 
teresting to learn when it was first used by authors; hardly, I fancy, 
before the middle of the fifth century, at the earliest. 

The form Istrahel (ff. 14 a, 15 a) is also very significant. It is 
the most ancient of all Latin spellings of this word, and was probably 
soon obsolete ^ 

The second declension form Danihelo (f. 14 a, 123 vb) is very 
ancient and most reassuring'^ (cf f 69 b). 

I should have little hesitation in assigning the archetype of our 
MS to the fifth or sixth century, and in view of its excellence, to 
Italy. Our codex is tentatively identified by Holder with no. 201 
in the catalogue of the Reichenau collection, which was written 
between 842 and 850, and is now preserved at Donaueschingen h 
Lest any one should doubt the probability of connexion between 
Reichenau and Italy, it may be pointed out that MS Augiensis LVii 
was written in North Italy in the second half of the eighth century ^ 
There are also other indications pointing the same way^. 

A study of the textual corruptions in the manuscript suggests 
the character of its archetype: 

(1) a for w: totam, asu, auditas, intellegant, factas, negaturam, 
instead of totum, usu, auditus, intellegunt, /actus, negaturum, 

(2) u for a: intellectu, erunt, for intellecta, e^'ant. 

(3) e for o: deuteresis bis, macedene, salomene, for deuterosis bis, 
macedone, salomone. 

(4) a for o: confundar, laboriase, for confundor, laboriose. 

(5) a for tu: confirmear, for confirinetur. 

(6) e for c: cheneris, for chencris. 

^ See Traube, Nomina Sacra, p. 106. It must, however, be remembered that the 
abbreviations of this word are very frequent and disguise the spelling of archetypes. 

2 See C. H. Turner in Jourii. Theol. Stud. vol. vi (1904—5) p. 253. 

3 Die Reichenauer Handschriften u.s.w., Bd. iii (1) (1916) p. 100. 

* Holder's article ' Der Isidorus-Codex Augiensis lvii der gr. Hof- und Landes- 
bibliothek in Karlsruhe' in Melanges qfferts a M. Emile Chatelain (Paris 1910), of 
which article I possess by Holder's kindness a copy corrected in his own hand. 

5 The Ambrose MS, formerly at Reichenau, now at St Paul in Carinthia, an 
uncial of the period between 400 and 700, was perhaps written at Verona (Die 
Reiehenauer Handschriften, Bd. iii (2) (1917) p. 127). Also, Codex Augiensis cix 
(saec. IX in.) was copied from a sixth century Beneventan original {op. cit. Bd. i 
(1906) p. 284). 


(7) u for o: bonus, -rus, nouatianus, for bonos, -I'os, nouatianos. 

(8) e for u: credentur for creduntur. 

(9) for u: mortuos for mortuus, infructoosi for infructuosi, 
uoluntariom for uoluntarium, obsequio for obsequiuni. 

(10) c for ^: cratiam, dilicatis, for gratiam, diligatis. 

(11) j!) for/: reputent for refutent. 

(12) r for 5: redderemur for redderemus. 

Most of these corruptions can be best explained by the use of 
a half-uncial archetype: (10), however, avouM occur more easily at 
the uncial stage. We may therefore assume that the immediate 
ancestor of our MS was a double-columned manuscript in half- 
uncial writing. 

A general study of the orthography of a manuscript like this 
leads one to certain conclusions as to the orthography of its original. 
If it be found that five pre-Caroline scribes agree in their spelling 
of a particular word, there is little doubt that the spelling they give 
was that of the archetype. There is in this codex a certain amount 
of consistency, combined with a certain amount of inconsistency. The 
following spellings, either because they are found everywhere in our 
MS, or because they are of such special excellence that they would 
not be known to the scribe apart from his exemplar, may be assigned 
with certainty to our Italian fifth- or sixth-century half uncial 

The usual final d in certain words appears almost everywhere 
as t: aliquit, aliut, aput, athuc^, illut, numquit, quitquit, siquit; but 
istud (probably because of the preceding t). 

Hiesus, thus written in full, is the form employed for 'Joshua* 
and for 'Jesus who is also called Justus': in contrast to the abbre- 
viated form, employed for 'Jesus,' as a sign of sanctity ^ 

suscrihere (twice), mercennarius, discidium, gluttire, balbuttir^e, 
ajluere^ (and derivatives), condicio (less often than conditio), solacium 
(oftener than solatium), haereticus (whereas most MSS have the less 

1 Cf. Dauit at least once. The form athuc is not mentioned by the Thesaurus. 

^ Traube discovered that these sacred names were abbreviated as a sign of sauctitj'. 
A confirmation, not mentioned, I think, by Traube, is to be found in the N.T. Greek 
minuscules 118 and 209, which in Matth. xxvii 16 and 17 write the name Jesus in 
full in the case of 'It/o-oOi' Bapa^^dv. 

^ See the notable article s.v. in Nettleship's CoJitributions to Latin Lexicography 
(Oxford, 1889). 


correct hereticus from kp€TiK6<i), susurrio, are perhaps the most 
notable cases of good ancient spellings. 

It is necessary now to summarise the orthography of MS A, 
which is not in all respects consistent throughout, and the first 
matter to consider is that of assimilation or dissimilation of pre- 

ad- adb. adbreiiiare. 

adc. adcommodare; but accipere (always). 

adf. adferre, adflictio, adfirmare etc., adfuit (always), but affectus, 

adg. adgredi, adgrauare; but agnosco etc, (always). 
adl. never; always allegere, alligai'e. 

adm. admittere always. 

adn. adnuntiare, adnectere always. 

adp. adpendere, adpetere, adprehendere, adponere, adprohare, 
adplicare; but appeUare. 

adq. adquirere etc.; never acq-, which perhaps did not survive 
the classical period. 

adr. adrogans etc., adridere; but arripei^e. 

ads. adsistere, adscrihere, adseiierar'e, adsignare, adsumere, 
adstruere, adsiuiilare, adsimidare, adserere, adstringere. 

adt. adtemptar^e, adtendere, adtestari; but attendei'e (generally). 

con- conl. conlaudare, conloquium; but colloquium, collocare, 
coligare (or -ere), collatus etc. 

conm. conmendare, conmonere; but conwieniorare etc., communi- 
care, commonere, commendare etc. 

conp. conplacere, conparare, conprehendere etc., conprohare, 
conpungere; but cortiparare, comprehendere, compar, comprohare, 
comperijx, compassio, compingere, coniplecti etc. 

conr. always assimilated, except in coni'egno: — corropoi'cu'i, 

in- (negative): inmundus etc., inlicitus, inmoderate, inmensus, 
inpudenter, inreprehensihilis (always), inremedmhiliter; but im- 
mundus, impiiis, impudicitia, irritus, im'inensus, immortalis. 

in- (preposition) : inrigare (always), inlatiis, inluminai'e, inritare 
etc., inmittere, inponere, inpendere, inligare; but impugnare (always), 
implere (always), impetrare etc. (always), impendere, impedire, 
imputare, irritare, immammillis. 

s. p. 14 


ob- off. (always). 

obm. (always). 

obp. ohproprium; but optemperare (very ancient). 

obt. obtulerunt etc. (obtare), obtundere; but (optare), optinere 
(very ancient). 

per- unassirailated, as far as used at all. 

sub- subportare, subponere; but suptilitas. 

exs- versus ex: exsistere; but expectare (always), expoliare 

On the whole the aspirate is correctly used in this manuscript, 
but we find exceptions, such as, in single instances: abentem, aesi- 
tationibus, losep, exortari, catecizare, arena, pasca on the one hand, 
and hisdem (twice), catkecuminus, hactu, Thimotheus (a very old 
spelling), Stliephanus (twice), habundare, thorns, liillis,abhominabilis, 
heriim, Thabitha, Machedonia, Honesimus (twice) on the other. 

Ae- and e- are confused in the following cases: 

-ae for -e: aequae, praemere, praetium, praetiosus, aepidari, 
c^lebrare, c§na, caena, speciae, depraecari, quaerella, caelare, celestae 
(bis), contrariaetas, uerae, praesbyter (usually), praessura, c^perant. 

-e for -ae: pe7iuria, -ate, enigma, lesio (twice), tedere, penitere, 
meror, eimdatio. 

t for c: audatior, mendatium, sotius, sotietas,pernitiosus,fallatia, 
iuditium, dilitiae. 

c for t: eciam, nupciae, quociens, uicium, infancia. 

y for i : Saphi/ra, Sapphyr'a, Helyas, elymosina, misteryo, cybus. 

i for y: praesbiter, Listra, elymosina, elimosina, misteryo, mar- 
tirium, praesbiteriiivi, azimus, idolothitus. 

b for p: deturbare, benetrare, ' 
probrius. These confusions are perhaps 

j9 for 6: plasphemare. i Irish. 

t for d\ inipetiuit. ] 

The reduplication of single consonants and the omission of one 
of a pair of double consonants may be regarded as an Irish, or at 
least an insular, symptom : 

(a) sabatum, presura, necesitas (bis), Philipenses, dificilis, positis, 
aparere, abysus, but sabatum and Philipenses need not be so 

(6) Paralippomenon, Corrinthii. 


suuam and eimangelium are not to be so regarded, however, as 
they appear in Italy already in the sixth century. 

Some spellings may be styled Merovingian, as they were frequent 
in that period, and were in great part cleared out by the Caroline 
reform. I refer to sterelis, agnusco (and cognusco), crededi (and such 
like: third scribe invariably), mercis {=merces, nom. sing.), exesti- 
mare, prumptus, recipissemus (and such like). 

The interchange of h and u occurs once or twice. 

The reader will find the orthography of this manuscript reflected, 
as far as possible, in my text. It may not be the orthography of 
Pelagius, but if it be not that, it is at least the orthography of the 
century succeeding his. For further particulars of the orthography 
of this and the other MSS, the reader is referred to the ortho- 
graphical index in the second volume. 

Orthography, particularly the orthography of proper names, and 
correctness of numbers, are perhaps the severest tests that can be 
applied to a Latin manuscript. This manuscript answers the former 
test well, and there is no opportunity to apply the latter. But in 
the fourth chapter we have already found that the text of the 
Biblical lemmata has been very thoroughly revised from the form 
which Pelagius used, and it may be that such revision has not 
stopped at the lemmata. Our manuscript was evidently copied, and 
carefully copied, from a clean manuscript, but behind that clean 
manuscript there must lie a rough copy into which harmonizations 
with the Vulgate had been inserted, while the original Biblical 
text was erased to make way for them. Or, if this was not the 
procedure, the scribe was instructed to keep his eye upon a Vulgate 
copy of the Epistles, placed in front of him simultaneously with the 
Pelagius, and to substitute the lemmata from the Vulgate for those 
which he found in his text of Pelagius. The substitution was not, 
however, completely made. 

A careful study of the textual differences between A and B is 
not always to the advantage of A. There are cases where A is un- 
doubtedly wrong; there are other cases where A is under suspicion 
of error. The differences between A and B, apart from the Biblical 
lemmata, are hardly of such moment as to suggest that we are 
dealing wdth representatives of two author's editions, as it were. 
It is safer to suppose that where A's latinity differs from that of B, 



and at the same time lacks the support of any other manuscript, 
we are in presence of alterations made by some early mediaeval 
reviser in the interests of what he thought was better Latin. But 
the text of A is, also, on the whole shorter than that of B. There 
are a few passages in B which are absent from A. Here again, we 
could hardly suppose two author's editions, since the passages are 
not numerous, were it not for the fact that the Vatican fragments, 
where they survive, lack the same passages as are absent from A. 
It would appear then that after all B does, in this particular, re- 
present a second (early) edition. In order that the reader may see 
clearly for himself what these passages are, I have caused all passages 
that are present in one of the two, but absent from the other, to 
be enclosed within square brackets. 

At the following points, among others, portions are absent from 
A that are present in B (and other authorities). 

In Rom. iiii 12, 16; vii 5, 6, 14\ 22; viii 26 bis; viiii 16; xiiii 2, 
23 bis; xv 4, 8; 1 Cor. iiii 12; vi 20; vii 4, 5; xvi 2 etc.; prol. 2 Cor.; 
2 Cor. ii 7 etc. 

In the following passages, among others, there are errors in the 
parts given by A. 

In Rom. viii 17. Here the lemma is given twice, first as ut et 
congloriJicemu7', second as ut simul glorificemur. The first form is 
due to our Vulgate interpolator, being the Vulgate reading. The 
second is the Pelagian reading, being that of DF*W d dem Lucif. 
Ambst. Anibr. etc., given by Wordsworth and White. 

In Rom. viiii 2. nee eum mendadi reum in aeterna accusatione 
constituit A, wrongly, for the true reading 'interna.' 

In Gal. ii 10. qui omnia sua distribuentes ad apostolorum pedes 
pretia deponebant A, wrongly, for the true reading 'distrahentes' 

In Gal. ii 12. ideo reprehensibilis erat quia se postea propter 
homines subtrahebat A. This is editing. The true reading is ' non 
ideo reprehensibilis erat quia cum illis edebat sed quia se postea 
propter homines subtrahebat.' Probably the five words were accident- 
ally omitted at some stage from homoeoarcton, and the non was 
then removed to make sense. 

In Gal. iiii 2.5. de qualitatibus locorum uidt intellegi diuersitatem 
1 Here the Vatican fragments join A. 


meritorum A, for the true reading 'testamentorum.' It is a sort of 

In Eph. i 17. sapientiam a domino deprecatur: nouerat enim 
€0.711 adiutricem omnium esse uirtutum A. This is a very interesting 
case: A is supported by Hi, but the true reading 'matrem' is given 
by BVG and is supported by 'matrimonium' of Hj {co7v. C). It is 
quite in accordance with Pelagius's attitude that he should use the 
stronger expression: it is equally natural that a reviser should tone 
it down. 

In Eph. i 18. si diuitias hereditatis dei uideretis, omnis terrena 
nobis horrebit hereditas A. The true readings are ' uideritis ' (fut. 
perf. indie.) and ' sordebit,' with B and other authorities. 

In Eph. iiii 7. qui ad quam gratiam se ajjtauerit, ipsam con- 
sequitur A etc. The sequence shows that the true reading is 'con- 
sequetur,' with V, 

In Eph. iiii 22. qui ptristinos errores desiderat huius viundi 
AB etc. The true reading, given by N, R m 3 etc., is ' deserat.' 

In 2 Tim. ii 26. non de dei bonitate dubitat, sed de accipientium 
uanitate A. The others read rightly ' prauitate.' 

(2) Codex Collegii Balliolensis Oxon. 1.57 (B) 

MS 157 (Arch. E. 5. 2), 217 (219) folia (1 column with 33 lines 
to the page), measuring 405 by 292mm., saec. XV med. in a beautiful 
Italian hand, of which fol. 15 r has been represented by photograph}^ 
considerably reduced in size\ 

1 flyleaf + 7 (= X 1 2 3 i 4 5 6 7) + 16 regular quaternions 

': I I I I I I I I 

+ 82 leaves with which we are not strictly concerned + 1 flyleaf. 

fol. 1 V. 'leronimus super eptsiolas omnes pauli praeterquam 
ad hebraeos.' 

' Hieronymus ' 'Liber domus de Balliol in Oxon ex done Willelmi 
Gray Eliensis epi.' 

fol. 2r. H 12 iudea : Eadem enim passi estis et uos a... 
manentem substanciam. 

1 Proceedings of the British Academij vol. vii (1916) opposite p. 17 of offprint 
( = p. 277). 




AD ROMANOS {red) (r)omani sunt QVI : 

ex iudeis gentibusque crecliderunt 

.,.(f. 2v) et ad concordiam cohortatiir. 


(f. 3 r) E^PISTOLA : AD : ROMANOS {red) 

(p)avlvs; Querimns quare paulus... 
(f. 134 v)...excipit apostolus quos saliitet. 


(C)ui apostolus a Roma...debeamus 
Paulus uinctus...(f. 135v)...fidei societate. 


(f. 136 r) after five lines' interval begins genuine Jerome on the 
Epistle to the Galatians, which is followed by genuine Jerome on 
the Epistle to the Ephesians, which ends this superb codex, Avritten 
in the most beautiful Italian style on the finest white vellum, with 
broad margins. It is unfortunate that in this, as in many another 
manuscript, the coloured initials were never filled in. At the very 
end occur these words:* 

Explicit liber beati hieronimi super gpistolas pauli apostoli. 

One leaf is missing from the beginning, containing most of the 
Primum quaeritur prologue. This appears to be the only part of 
the MS itself that has perished, but we shall see later that at least 
two considerable portions of the archetype had perished before our 
copyist took it in hand. 

The orthography of a manusci'ipt like B shows three character- 
istics. It shares the orthography of its date, what may be called 
the humanistic or renaissance orthography. There are also in certain 
words affectations of antiquarianism such as some of the scribes of 
the latter part of the fifteenth century display. But, thirdly, there 
are certain undoubted relics of the orthography of the archetype. 
It is only these last that are of special importance to the editor, but 
for the sake of completeness we must take account of all three styles. 
I have caused those spellings which I think may be attributed to 
the archetype, to be printed in thick type: the affectations of anti- 


quarianism are represented in italics. Most of the spellings are 
recorded in the orthographical index, and are thei*efore not repeated 

spiritalis, quicumque, saltim, Arrius, Fotinus, holus, cotidie, 
Sarra, littera, oblitterare, secuntur, eundem, Eleazarus \ zabulus, 
obprobrium, caelum, inmundus, inmensus, heremus, Philipus, 
Philipenses, Tessalonicenses, Colosenses, adtendere, discidium, 
Sostenes, ammirari, lacrima, inmunditia, scisma, cena, opor- 
tunus etc., carisma, commendaticius, quotienscumque, am- 
ministratio, abicere, coniti, inreprehensibilis, coartare, Grecus, 
obaudientia-, obauditio, obaudire, adsignare, inpetrabilis, in- 
pendere, arte (adv.), eicere, cohercere^ quicquam, hereditas, 
subplere, suplere^ conmilito (noun), allegere, alligare, ammo- 
nere, conlaborare, Filetus, conregnare, mammilla, locuntur, 
coniunx, tem.ptare, ualitudo. 

quotn, quoins, dissentio (noun), -isare. The very ancient forms 
quom and quoins, which were of course never employed by Pelagius, 
are an affectation of the scribes of this period, the second half of 
the fifteenth century ^ It was a classical, not a theological scribe, 
who copied this MS to the order of Bp Gray, who visited Padua, 
Ferrara, and Florence, and who, when he could not obtain a manu- 
script, was able to pay a professional scribe to copy it^ 

Assimilation of prefixes is the rule: I think all the exceptions 
are in the list just given: ti and ci are both found as representatives 
of the assibilated sound, hut fatiunt, /cities etc. are almost invariable: 
i is often found for y, and y for i: we usually find the intrusive p 
in such cases as condempnare ; even in uerumptamen : marcesso 
occurs for marcesco. All these may, I think, be regarded as examples 
of the average fifteenth century spelling. 

1 Where A has Lazarus ( = vg). 

- The scribe here betrayed himself by writing this word as two words: if he 
had imagined that it was another form of oboedientia, he would have written 

>* So in archetype : corrupted to coherere. 

■* Cf. the Verona li (49) uncial MS of Maximus of Turin etc. ed. C. H. Turner 
{Journ. Theol. Stud. vol. xx [1918—19] c. 6 1. 13 (p. 301), c. 11 1. 38 (p. 310)). 

5 Cf. A. E. Housman io the Journal of Philolofiy, vol. xxi (1893) p. 180 n. 1, 
confirmed by A. C. Clark in a private communication to E. A. Lowe, whom I con- 

6 See the article on him in the D.N.B. 


There is clear proof that the archetype of our MS was in insular, 
probably Irish, script, and it is an easy conjecture that this archetype 
was a Bobbio manuscript, Bobbio being the nearest place where 
such a manuscript was likely to be found. 

The Balliol MS passed through an English hand, to which are 
due the 'non bene' in the margin opposite in Rom. v 4, where also 
an error is emended by the same hand; the 'quo sola fides sufficiat 
xpiano ' in the margin of fol. 82 v ; also headlines such as ' 2 C02+ ,' 
and the occasional addition in the margin of the numbers of modern 
chapters. It is interesting to note that 4 = quod, while ^ = quid: 
sometimes errors are made in this connexion. A faint cross is often 
put over words that are corrupt in the manuscript (e.g. in 2 Cor. 
xiii 11). 

The archetype was perhaps in double columns, each about 19 
letters broad. This is suggested by the repetition of membra 
iimguntur after the second corpus in the comment on Col. ii 19. 
The manuscript seems to have been at times illegible through age 
or exposure: for not infrequently one or two words are omitted. 
Sometimes the omissions are more serious. The Balliol MS is thus 
written at in Rom. i 13 (f. 4r): 

Non enim arbitror ignorare uos fratres quia saepe proposui 
uenire ad uos. Per commeantes enim fratres audire potuistis: 

et proi 
hibitus sum usque ad hue: proibitus hie: occupatus accipitur 

Ut aliquid fructum habeam et in nobis... 
Similarly at in Rom. i 18 (f. 4v): 

ex operibus legis. reuelatur ira dei de cglo. Incipit ad partem 

nouerunt enim homines et benefitia et plagas expectare de caelo : 
in om 
There is no doubt as to the meaning of this. On both sides of 
one leaf of the archetype there was a stain obliterating the equi- 


valent of two and a half lines of writing of our MS, or a portion of 
that size had actually disappeared from the MS altogether^ The 
scribe of the Balliol MS has been very careful to represent the 
amount of the loss in each case by a blank such as is reproduced 
above. But these are not the only instances. Another case occurs 
later in the manuscript. 

At in Rom. xi 28 (f 28 v) the Balliol MS reads thus: 

bis predico xpra. Prohibentes nos 

quia deum non penitet abrahae semini promisisse : siue illi sine af- 
and at in Rom. xi 33 f (f 29 r) thus: 

cogitatio dispositionum eius. Quis enim cognouit sensum domini 

aut quis eius consiliarius fuit ut eius nouerit archana sacramenta. 

This lacuna, which is rather longer than the last, is to be ex- 
plained in the same way. The scribe of the Balliol MS was once 
again very careful to represent the extent of the loss. 

It is quite evident that the archetype of our MS was in a de- 
fective condition. But the losses just pointed out are trifling in 
comparison with others which have to be mentioned. 

In Rom. xii 17 'maxillam praebere — iii Rom. xiii 12 abiciamus 
simul ' had disappeared from the archetype, without leaving any sign 
in the Balliol MS. Here it is a matter of leaves, not of lines. 

In 1 Cor. xi 28 'probet autem se ipsum — in 1 Cor. xv 3 secundum 
scripturas (alt.)' had also disappeared from the archetype, but here 
the Balliol scribe has written the word 'de est' at the beginning of 
the omission. A computation will show that this second omission 

1 Perhaps the best known instance of such a stam (and its consequences) is that 
of Cod. Bodl. Gr. Misc. 251 fol. 25 a, of Arrian's Epictetus (see the photograph in 
H. Schenkl's edition). 


is four times as long^ as the first. It is easy therefore to conjecture 
that in the second case a whole quaternion of the archet3"pe had 
disappeared, and in the first case the two inmost conjugate leaves 
of another quaternion. 

It is possible to tell more about this defective archetype, from 
a study of the errors in the Balliol MS. 

n is written for u: some case of noster for the same part of uester 
(in the archetype of course in the contractions nr, Tir, ni, ui or nri, 
uri, so that there is really no confusion of o and e) on flf. 31 r, 57 r, 
68 V, 69 r, 72 v, 79 r, 86 r, 94 r, 99 r, 105 r; nos for uos on ff. 39 v, 
43 r, 49 r, 62 v, 73 r bis, 74 v his, 84 r, 90 r, 112 r; nobis for uobis on 
ff. 79 r, 82 V, 119 v; ne for ue f. 93 v, ant for aut f. 116 r, sine for 
sine £ 54 v, angures for augures f 26 v, noliint for uolunt flf. 88 v, 90 r, 
noluerit for uoluerit f. 49 r, boni for boui f 50 v, nouit for uouit f 109 r, 
senior for senior f 111 v, anniculo for a uinculo f 49 r, ioninianum 
for ioninianum f. 73 r, cf molatur for iiiolatur f 99 v, praemiiun for 
praeuium f. 112 v, amore for aiuniore: i.e. 34 (37) cases. 

u is written for n: nos for nos flf. 50 v, 57 v, 113 r; nobis for nobis 
ff. 61 r, 67 r, 74 v; case of uester for case of noster f 106 r: aliquo 
for alieno f 41 r, peruitiosae (or per nitiosae f 84 v, iamues for iamnes 
f 112 V, n<s^a for zns^a f 124 r, auiis for anus f 124 v, autem for anie 
£ 132 r, sine for szne on ff. 47 r, 50 r, 56 r, 127 r, diuiuani for diuinani 
£ 59 r, ammoueri for amnioneri £ 109 r, commouet for conimonet 
£ 104 V, commouentur for comnionentur £ 124 v, scandalizaidt for 
scandalizant £ 41 v, euacuauit for euacuant £ 38 r : r.e. 23 cases. 

u is written for a: morabuntur for niorabantur £ 8v, exierunt 
for exierant £ 24 v, poterunt for poterant £ 31 v, c?zt?7i querere for 
damnare f 32 v, .smzo?i for swam f. 33 r, bonum for bonam £ 33 r, 
crediderunt for crediderant £ 34 v, seruatum for seruatani £ 44 v, 
cognoscunt for cognoscant £ 47 r, diuinuni for diuinani £ 49 r, dicuntiir 
for dicantur £ 49 v, terreus for terreas £ 74 v, solum for solam £ 84 v, 
edificandum for aedijicandam £ 94 v, e»?n for eamf 97 v,guom (=cum) 
for caro e^ £ 98 v, panlus for />?'o oii/s £ 115 v, /iie/^ii^if for fuerant 
£ 133 V, secundum for secundam £ 134 v, corrigunt for corrigant 
£ 134 v: 20 instances. 

^ At tbis stage I have only the copy of the text I wrote out for the printer to go 
by : in the first case the loss is represented by 105 lines of my writing, in the second 
by 421 ! 


a is written for u: ueram for uerum f. 11 v, {illas for illius f 14 v), 
fractus for fructus £ 17 y, factum for fructum f 35 r, dormierant for 
dormierunt f 56 v, agant for agunt f. 60, mandanda for mundanda 
f. 69 V, perierant for perierunt f. 96 v: 8 (9) instances. 

r is written for s: sustinemur for sustinemus f 21v, pro porci 
for propositi f. 31 r: 2 instances, 

s is written for r: collaudenius for coUaudemur f 33 r, liberatiis 
for liheratur f 35 r, c?eMS ((/s) for dicitur (dr) f 42 r, separ'are for 
reparare f 58 r, prodest for prodere f. 77 r, ignorantes for ignoranter 
f 77 V, conuersemus for conuersemur f 87 v: 7 instances. 

t is omitted at the end of the third person singular and 
plural of tenses of verbs : diligi for diligit f 3 v, iyipone for 
inponet f. 19 v, coinpleui for compleuit £ 39 r, inoyses for mo5 esi 
£ 43 V, admitti for admittit £ 48 r, e,r causa for excusat £ 61 v, j97'o- for prodesset £ 81 r, reuocare for reuocaret £ 92 r, (audere for 
audent £ 100 v), sin for sin^ £ 108 v, (terrena for terreant £ 131 v), 
pertinen for pertinent £ 131 v: 9 (11) instances. 

r is written for n: redemptorem for redemptionem £ 10 v, arserit 
for manserit £ 41 v, audere for audent £ 100 v; c£ mittere for mitem 
£ 130 v: 3 (4) instances. 

n is written for r: colentur for coleretur £ or, uidentur for 
uideretur ff. 84 v, 88 v, 97 r, monitmxim for morituram £ 116 v, 
(quantitatis for caritatis £ 106 r), c£ sedem for sedere £ 91 r, iw- 
sectarem for insectarer £ 103 v: 5 (8) instances. 

r is omitted after or before a vowel: deseuio for deseruio £ 3 v, 
uetere for uertere £ 28 v, suppotauit for supportauit £ 33 r, factum 
for fructum £ 35 r, pe for_^3er £ 125 r: 5 instances. 

r is added after a vowel : hortandumi for notandum £ 30 r. 

d is written for t: c/e for te £ 131 r. 

t is written for d: sanctum for secundum £ 3r (see below), 
conteiiiptamus for contendamas £ 32 v. 

c is written for t: jaro 2)orci for propositi £ 31 r, (adduci for 
adducti £ 33 v), joftce factum for patefactum £ 36 v, decrimento for 
detrimento £ 44 v, idolaticis for idolotitis £ 49 r, circa lor ierra £ 58 r, 
cerfe^ for tedent £ 6i r, stc for siY f. 61 v, archeman for artheman 
£134v: 9 instances. 

t is written for c: si^ for sic £ 19 r. 

c is written for g: aucmentatur for augmentatur £ 100 v. 


nt is written for m: sunt for sum f. 38 r, manifestent for mani- 
festem f. 119 r. 

m is written for nt: hesitabam for haesitabant f. 59 r. 

a is Avritten for o: coartar for coarto7' f. 101 r. 

h is written for n: hortandum for notandum f. 30 r, barhabas 
for barnabas f. 44 r. 

li is written for h : liuius for huius f 43 v. 

t is wn-itten for h: tunc for hunc f. 35 r. 

1 is written for i: lucidia for iudicia f 29 r. 

c is written for e: qiiom (cum) for eum fF. 43 v, 77 v, 112 v, aliquo 
(alicuo) for alieno f. 41 r, cf coherenda for cohercenda f. 87 v. 

e is written for c : diem for dicit f. 79 r (t.e. die for die), asn^e- 
ritiim for asincritum f. 36 r, aerore for ac{r)ore f. 43 v, ei(??i for c2/?>i 
f 55 V, ezeg for eice f. 86 r, montem eum for monte cum f. ] 01 v. 

u is written for ti: euam for e^ia»i f. 50 v. 

u is written for g: leuatione for legatione f. 68 r. 

s is Avritten for f: sit for fit f. 96 v, apocrisis for apocrifis f 112 r. 

f is written for s: yic?es for si des f, 31 r, fulcis for sulcis f. 7o r, 
inutile fiant for inutiles fianti. 9H r, {inutilifuitf or inutilisf ait f. 135 v), 
per infima for peripsima f 42 v. 

o is written for a: amotorem for amatorem f. 109 r, longuescens 
for languescens f. 126 r. 

a is written for o : idolaticis for idolotitis f. 49 r, sodamitis for 
sodomitis f. 96 v. 

a is written for it: latei'is for litteris f. 115 v. 

n is wi'itten for ti: ambulans for ambulatis f 40 v, permanens 
for permanetis f. 115 r. 

r is written for t: oblitterai'um for oblitteratum f. 14 v, uidere 
for uidete f. 59 r. 

t is written for r: reddat for reddar f. 119 v. 

The following may be classed together as confusions of letters 
in cases where these letters consisted mainly of short vertical 
strokes^; these suggest that the cross-strokes in the archetype were 
very faint: 

petiuit iov petunt f. 38 v, uestiuit for uestiunt f. 12^v, finiunt for 
finiuit f. 35 v, py^ojicuit for projiciunt f. 95 v, dubitantuit f. 103 r. 

in for ut f. 46 v, 121 r, 133 v. 

1 See also uuder deus, doniinus in the following list. 


inihant for nubant f. 46 v. 

uni for uim K 38 r, 53 r, I04*r, tribuni^ for tribum f. 83 v-, ti^o- 
phnnum for trophimuvi f. 132 r. 

peremptus for penit us i 6'2 v. 

seruam for seriuiui f. 75 v. 

missum for iussum f. 98 v. 

iouimanistis for iouinianistis^ f. 104 v. 

m for Ai f. Ill V. 

camis for carius f. 112 r. , 

mutari for imitari f. 113 v. 

wi for m flf. 36 r, 123 r. 

Other eri'ors are due to ignorance or mishandling of abbrevia- 
tions in the archetype: 

sanctum for secundum (if) f. 3 r. 

gloriam for gratiam^ (gra interpreted as gla) £f. 13 r, 61 v, 66 r, 
71 r, 79 V. 

gratia omitted (gra again not understood) f. 118 v. 

autem omitted (because in the form Ir) ff. 14 r, 48 v, 64 v, 75 v, 
84 V bis, 110 V, 123 r, 125 v, 126 r, 131 r. 

enim omitted (because in the form tt) ff. 78 r, 99 r, 101 r, 113 v, 
124 r, 126 r, 133 r, 134 v. 

enim written for autem {i.e. tt read for h*) tf. 16 r, 34 v. 

eMi'm for ergo {i.e. tt for g) £ 57 v. 

autem written for enim {i.e. h* read for tt) ff. 42 r, 64 v, 78 v, 119 r. 

igitur omitted (because in the form g') f. 14 r. 

igitur written as ergo (because g' taken as g) ff. 94 v, 103 r. 

ergo omitted (because in the form g) f. 14 r-\ 

deus written for dicitnr (because dr confused with ds) f. 42 r. 

secundum written for saeculum (sclm being in the archetype*) 
ff. 48 r, 52 r. 

1 These errors, combined with other facts (see p. 215 above), suggest that the 
scribe of the BaUiol MS was a non-ecclesiastical scribe, accustomed to the copying 
of pagan texts. 

- See also under deiim etc. on p. 222. 

3 In cases where autem, enim etc. are interchanged in Biblical lemmata, it would 
not always be safe to assume error on the part of the scribe : the difference may be 
sometimes textual rather than palaeographical. 

^ The ignorance of the saeculum contraction supports the contention in the note 


epistolas written for epulas (because he thought he saw eplas 
in front of him) f 55 r. 

deuvi for domini {i.e. dm for dm) ff. 55 v, 79 v, 100 r, 103 r. 
• dei for domini {i.e. diii confused with di) f 118 v. 

dominus for deus {i.e. dns for ds) f 57 r. 

uult omitted (perhaps because written as uul'^) f 59 r. 

peremptus for penitus is perhaps a dittography, f. 62 v. 

uerhis is written for nobis {i.e. uB was taken to be lib) f. 63 r. 

tanturti written for tamen,{i.e. tn taken for tm) ff. 66 v, 115 r. 

eius omitted (because in the unintelligible form ^) ff. 67 r, 116 r. 

ministrat written for viinistratur {i.e. the ' or - above the second 
t unobserved) f. 72 r. 

per written for pro {i.e. p confused with p) flf. 74 v, 105 v. 108 v, 
133 r bis. 

pro written for per {i.e. p confused with ^) f 101 v. 

quia written for qua{m) {i.e. g, taken as q.) f. 75 v. 

uel omitted (because in the form I) f. 80 v. 

quod written for quia (because g. misunderstood as q.) f 80 v. 

est omitted (perhaps because in the form -f-) f 81 r. 

quom {cum) dat written for condat (because odat was treated as 
cdat) f 92 r. 

aut written for a (the scribe mistook the accent over the a as 
the abbreviation sign : i.e. he took a for a) f 98 v. 

homo omitted (because in the form ho) f 102 r. 

esse omitted (perhaps because in the form ee) flf. 38 v, 134 r. 

aduersus written for ad usus (perhaps the scribe had aditsus 
before him) f 134 v. 

nostram omitted after formam (perhaps because in the form 
nam, which would make haplography easy) f. 60 r. 

No one who has studied Latin palaeography will, I think, doubt 
where all this evidence points. The manuscript which the Italian 
scribe of our Balliol MS set out to copy, was in Irish pointed script 
not unlike that of the Book of Armagh to which it is textually 
related. As has been already hinted, this manuscript had probably 
been at one time in the library of St Columban's Irish foundation 
at Bobbio. Certain of the corruptions mentioned above, such as u 

1 This form is at least as old as the sixth century, and doubtless comes from an 
ancestor of the immediate original of the Balhol MS. 


for a, s for r, c for t, may very well be inherited from a still earlier 

stage, the half-uncial stage. Another must, I think, be assigned to 

that stage, the confusion between m and nt; for this occurs most 

naturally in MSS where the transverse stroke following the vowel 

does double duty, siiT being alike sum and sunt, and this usage 

seems to be unknown outside uncial and half-uncial scripts 

The descent of the Balliol MS may then, I venture to think, be 

thus sketched: 

MS in half-uncial (saec. v — vi) 

. . I . 

MS in Irish pointed minuscule (saec. ix in.) 


(Italian minuscule [saec. xv med.]). 

The reader will already have derived some knowledge of the 
quality of B, from what has been said about A, but as an impression 
may have been created that B is really superior to A elsewhere, as 
well as in Biblical text, some instances may be cited to show that 
B is on occasion inferior to A. 

B interpolates the Marcionite prologue to First Corinthians. 

Again, at 1 Cor. x 22 A is right with ipsi me zelauerunt in non 
deo (Deut. xxxii 21 avTol Trape^rjXoya-dv /u,e eV ov Oeo)), against B 
with ipsi me zelauerunt in domino. 

Further, at Eph. iiii 13 A has in iinitatem fidei rightly, while B 
has homines sicut me. 

Other instances will readily be found by turning over the pages 
of the text. 

(3) Codex Collegii Mertonensis Oxon. 26 {saec. xv) (0) 

The manuscript numbered 26 in the collection of Merton College, 
Oxford, the gift of a former Warden, Richard Fitzjames, Bp of 
Chichester^ (1504 — 1506) is, from the multiplicity of its contents, 
perhaps the most wonderful Jerome manuscript in existence. It is 

^ It seems to be very little known : see my edition of Pseudo-Augustini Quaes- 
tiones (C.S.E.L. vol. l [1908]) p. xxx n. 3, and add Codex Bezae (Rendel Harris in 
Texts and Studies ii 1, 121 ff.), C.S.E.L. lix p. Ixv, andCassiodorus ( = Pseudo-Prima- 
sius) (Migne P.L. lxviii, p. 587 11. 44 — 5). 

- As fol. 5v records: as he is there called *uuper Cicestriensis epi,' the inscrip- 
tion may be assigned to the date 1506 or 1507. 


true that there are many coeval manuscripts containing large col- 
lections of Jerome's letters \ but to the best of my knowledge, there 
is no other single MS containing so many commentaries of Jerome 
on the New Testament. If I mistake not, it contains every genuine 
commentary of Jerome on a New Testament book as well as every 
New Testament exposition falsely attributed to him. Dom Morin 
elicited from it the genuine 'De Monogramma Christi,' which he 
published in 1903-, and it has been used for Dr Haussleiter's edition 
of the works of Victorinus of Pettau, so far as concerns Jerome's 
revision of the latter's commentary on the Apocalypse^ It has been 
left to me to study the commentary on fourteen Epistles of Paul 
contained in it, falsely attributed to Jerome. 

The manuscript 26 (B 3, 4) saec. xv (English hand) has on 
fol. 74 r the following title in red: 

Hieronimi presbiteri breuiarium incipit feliciter; then in black: 

Primum queritur quare post euangelia. . .meliorem et manentem 
substanciam; then in red: 

Explicit prologus omnium epistolarum beati pauli apostoli. Item 
argumentum solius epistole ad romanos; then in black: 

Romani sunt qui ex iudeis pacem et ad con- 
cordiam cohortatur; then in red: 

Explicit argumentum epistole beati pauli apostoli ad romanos. 
Incipit explanacio sci hieronomi (sic) in epistola ad romanos; then 
in black: 

Paulus. Querimus quare... gloria honor in secula seculorum 

(red) Explicit epistola ad romanos incipit argumentum prime 
epistole ad corinthios. 

(black) Corinthii sunt achaii...scribens eis ab epheso. 

(i^ed) Explicit argumentum incipit epistola prima ad corin- 

(black) Paulus quod nomen preponit... super tuam benedictio- 

1 I made an inventory of the contents of all, or nearlj' all, MSS in Britain con- 
taining letters of Jerome for Dr Isidor Hilberg of Czernowitz, the Vienna editor. 

2 Anecdota Maredsolana vol. in pars 3 pp. 194 ff. 

3 The collation was made by the present writer : see Victorini episcopi Petavio- 
nensis Opera. ..rec. Joh. Haussleiter ( = C.S.E.L. 49) (Vindob. 1916) pp. Ixix — Ixx. 


So with 2 Cor., Gal., Eph., Phil., 1 Thess., 2 Thess., Col., 1 Tim., 
2 Tim., Tit., Philem. 

(red) Explicit Epistola ad philemonem. Incipit argumentum 
ad hebreos. 

(black) Inprimis dicendum cur apostolus... greco sermone con- 

(red) Explicit argumentum. Incipit Epistola ad hebreos feli- 

(black) Multipharie multisque modis....Per multos inquit pro- 

Expliciunt annotaciones super epistolam ad hebreos (fol. 141 v). 

Then three blank leaves: 

Then genuine Jerome in Gal. Eph. Tit. Philem., which end the 

That the portion of this manuscript down to the end of the first 
commentary on the Epistle to Philemon, is a copy of the Balliol MS, 
immediate or mediate, is evident from the following facts: 

Its text agrees with that of the Balliol MS throughout, and 
this is true of no other existing MS; but absolutely irrefragable 
proof of the descent is furnished by the lacunae referred to above 
in the account of the Balliol MS\ 

The first lacuna, where two lines and a portion of a third are 
carefully left vacant in the Balliol manuscript, is represented in the 
Merton manuscript by the blank of half a line only. As we know 
what the missing words are, it is clear that the Merton manuscript 
is secondary to the Balliol manuscript. 

The second lacuna, also of 2 + lines in the Balliol MS, is repre- 
sented again by half a line in the Merton MS. 

The third lacuna is three and a half lines long in the Balliol MS» 
but the scribe of the Merton MS was content to indicate the gap 
by about a fifth of a line. His vellum was more precious to him 
than was that of the Italian scribe. 

The fourth lacuna measures exactly three lines in the Balliol 
manuscript, but only half a line in the Merton manuscript. 

The Merton manuscript is therefore secondary, and in con- 
sequence is not employed in this edition, except for the text of 
the general prologue which has almost entirely disappeared from 

1 pp. 216 f. 

S. p. 15 


the Balliol manuscript owing to the loss of its first leaf. Fortunately 
the Merton MS was written before this loss occurred, and here takes 
the place of the Balliol manuscript. 

(4) The Vatican Fragments (M)^ 

The two Vatican leaves belonged to a MS in half-uncial writing 
of about the sixth century, which was taken to pieces, at least as 
early as the eleventh or twelfth century, to make guard-leaves. No 
doubt the manuscript was already defective before it was delibe- 
rately broken up. The leaves are mutilated, scribbled over and cut 
down, and some nineteenth-centur}- bookbinder did not improve 
their condition by separating them unskilfully from some unknown 
MS or printed book. 

The leaves are conjugate, but not consecutive. They are in feet 
the third and sixth leaves respectively of a quaternion, thus: 

(3) (6) 

(1) (2) fol.l (4) I (5) fol.2 (7) (8) 

The vellum is good, if somewhat coarse, and has been ruled with a 
hard point on the softer and whiter side, both horizontally at even 
intervals and perpendicularly to govern the commencement and the 
course of the lines of wi'iting. On the external margin of fol. 1 some 
of the punctures still remain, which were placed there to make the 
lines equal. 

The margins and some lines of writing being absent, it is not 
possible to fix the exact size of the leaves in their complete state. 
Dr Mercati would reckon the written part of the page as approxi- 
mately 20 X 12 cm., and the whole page as at least 25 x 1.5 — the 
dimensions, that is to say, of a good-sized MS in octavo. The exist- 
ing fragments are unequal, and measure roughly speaking the one 
177 X 118 mm., the other 178 x 140 mm. 

The writing is regular and compact, and keeps to the horizontal 
lines: both from its own size and from the amount of space left 
blank between the lines it gives the impression of a certain richness 

1 See also chap, ii pp. 48 ff. I am almost entirely indebted to Dr Mercati's 
description (J.T.S. vol. vm [1906—1907] pp. 529 ff.). 


and magnificence. There are no initial letters larger than the rest, 
but at every new section — and a new section begins with every new 
stichos of the apostolic text to be commented upon^ — a commence- 
ment is made two letters outside the line, and the same in every 
succeeding line (except in fol. 2 r line 19, 2 v, 1. 8) until the lemma 
from the sacred text is concluded. Thus the text stands out at the 
first glance: and to ensure this effect two perpendicular lines are 
ruled down the page, one to serve for the beginning of the lines of 
the lemma, the other, further in the page, for the lines of the com- 
mentary. From this method of emphasizing the text, it results that 
part of the preceding line is sometimes left unoccupied ; and, con- 
versely, the line is, in such cases, sometimes prolonged further than 
usual, and the letters made smaller than usual, in order to end off 
the comment, as can be seen on fol. 2r line 17, without beginning 
a new line. In filling up the lost ends these unequal contents of 
different lines must be borne in mind: the lines vary from 23 letters 
(or even less) up to 36, the average being about 26. 

Punctuation by the first hand is rare, and is distinguished by 
being placed neither at the top nor at the bottom, but towards the 
middle of the letters. Abbreviations too are rare, and only the most 
common occur: ds dns sps xps sea, and at the end of the line a 
stroke for n and m^. q, for -que does not occur. One single ligature, 
the well-known one for -unt, is found on fol. 2 v line 20. Noteworthy 
divisions of words are subsjtantia, consjtructus, fol. 2 r lines 5, 23. 

Between the recto and the verso of each extant leaf three whole 
lines and part of a line have been cut away, so that the whole page 
consisted of twenty-six lines. Whether the lines which have com- 
pletely fallen out stood at the foot or at the head of the page, cannot 
be stated with certainty^ As has been pointed out above, the two 
inner conjugate leaves of a quaternion separated the two leaves 
which have survived, and these two inner leaves had the flesh sides 
inmost, as was usual. 

1 This fact is very important as testifying to the original arrangement of text 
and commentary: cf. also above, p. 50. 

- For m besides the stroke there seems to be also a point — , fol. 1 v, line 2: but 
the point may be one of punctuation, and not part of the abbreviation of vi found 
in other MSS. 

^ At the top of fol. 1 v Dr Mercati seemed to make out the tail of some letter iu 
the preceding line. 



Two later hands dealt with the MS. One of them, who used a 
dark ink similar to that of the text, is practically contemporary with 
the original scribe: he added dots for punctuation, placing them 
level with, or even above, the tops of the letters, and in four instances 
corrected or supplemented the text (fol. 1 r lines 4, 12; 1 v line 5; 
2 V line 15), probably from a second MS which in two cases gave 
the readings of the Migne text of Pseudo-Jerome. Whether ^ve 
should attribute to this or to the first hand the erasure in eg*o 
(doubtless egeo was written at first), fol. 1 v line 12, is not clear. 
The third, somewhat later, hand has confined its activity to marking 
the beginning and ending of the verses of the Apostle with a big 
stroke something like a bracket, making use of a dirty sooty ink. 
This hand's work can safely be neglected. 

Finally, after the two leaves had been already taken from the 
MS, various uneducated hands have touched up the ink of letters 
here and there, and have scribbled roughly across the page letters 
of the alphabet and the words probatio penne, proba, probatio in- 
constri, ave cuius (saec. xi — xii ?). One of them, in between the 
lines but upside down, has signed himself ' Ego dns (the profanity 
of the man!) adobad' cleric' plebis valliis (or 'vallus') renovata ' (sic). 
The name of the valley in question is quite unknown either to 
Dr Mercati or myself^ 

The fragments contain text and commentary for Rom. vii 9 — 15 
and viii 3 — 7, with certain slight gaps which are recorded in the 
critical apparatus. Short as these fragments are, they are most 
welcome; first, because they show the way in which Pelagius ar- 
ranged his work: second, because they give a Biblical text which 
is not Vulgate, which is in fact distinctly more Old-Latin than that 
furnished by the Reichenau MS: third, because, while their Biblical 
text is nearer to that of the Balliol MS than to that of the Reichenau 
MS (cf Rom. vii 13, 14, 15), the fragments agree with the Reichenau 
MS in omitting at Rom. vii 14 a portion of exposition which is 
furnished by the Balliol MS. 

For these reasons the Vatican sixth century fr-agments are, so 
far as they extend, the most valuable testis we have for the text of 
Pelagius' work. 

^ I regret very much that I omitted to take the opportunity of consulting the 
late Dr H. M. Bannister on this point. 


(5) The Freiburg Fragments (K) 

Two conjugate leaves of a Latin manuscript, discovered by 
Dr Flamm of the Stadtarchiv at Freiburg im Breisgau, Baden, 
came under the eyes of Professor J. M. Heer there, who most cour- 
teously forwarded rotographs of them to me in 1912. The leaves 
had been in use for many years as the cover of the accounts of a 
professors' gawk club^ at Freiburg, and are now preserved in the 
Stadtarchiv there. 

The writing may be safely assigned to the first quarter of the 
ninth century, and to a South German scriptorium'-. Certain of its 
palaeographical characteristics may be mentioned. Open a is in- 
variable. The diphthong ce is never so written ; generally it is re- 
presented by e merely, once or twice by ^. In the last lines the 
following letters are sometimes provided with long tails, reminiscent 
of legal documents: f, p, q, r and s. The short stroke is sometimes 
used both for m and for n, both in the middle and at the end of a 
word or line. The letter r is sometimes highbacked, and a low y is 
characteristic. The y is of rather peculiar shape and is dotted; the 
z is short, and stretches half below the line. The following ligatures 
occur : ec, et, ex, nt and ri. Punctuation is rare, and is of three kinds : 
the simplest is that of the dot placed half-way up above the line, 
the most emphatic is that of two dots in a line and a comma mid- 
way between them underneath, while the medium strength is re- 
presented by the semi-colon. The interrogative sign is entirely 
absent, though there is at least one place where the sense is inter- 
rogative: the absence of the interrogative sign is unfortunate, as 
its presence would greatly facilitate an answer to the question of 
the provenance of the MS. The presence of scripture quotations 
in the text is sometimes indicated in the margin in the usual way. 
The MS is carefully and neatly written; there is no attempt to save 
space. The separation of words is fairly advanced. The number 

^ Gesellschaft zum Gauch, stupidly rendered by me 'goose-club' in Journ. Theol. 
Stud. vol. XIII (1911 — 1912) pp. 515 — 519, where I published a diplomatic text of 
the contents of the leaves. The leaves enclosed vol. 55, the accounts of the club for 
the year 1592—1593. 

2 I was fortunate in afterwards securing Dr Alfred Holder's agreement with this 


XVI, indicating a capitulum, appears at 1 Cor. x 25. This is a matter 
of some importance, as 1 Cor. x 25 is regularly the beginning of 
capitulum L or LI: the question of what system of capitulation is 
here used must be left to experts. The scribe was somewhat addicted 
to the omission of syllables — euan{ge)lizare, ce{te)ris, idol{i)o, irn- 
mola{ti)cium, and he was not always sure of his vowels — potet for 
putet, and parte cipo for pariicipo. 

The following are all the abbreviations and contractions that 
occur: au, dm, dni, e, ee, frs, n, p, p, ppter, q; (= quae), qd, sea, xpi. 
The following syllabic suspensions occur : b; (= bus), t (= ter), 
t' (= tur), at end of line. 

The contents are text and commentary for (a) 1 Cor. viiii 15 — 17, 
{h) 18—20, (c) 1 Cor. x 24—27, {d) 27— 31, with lacunae equivalent 
to six lines of MS, first between {a) and (6) and second betw^een (c) 
and {d). That is, six lines of waiting have been clipped away from 
the tops of the four pages right across. As the pages now contain 
17 lines each, they must have contained originally 23. A calculation 
of the matter lost between (6) and (c) shows that eight pages are 
missing, and that our leaves are therefore the second and seventh 
of a quaternion. 

We must now consider the textual quality of the fragments. 
There are four cases where they go with A and B combined: 

potero immutare K'B>K.{Gi)\ poterani mutare V : potest immutariH.. 

exemplo apostoli ABKG, (-um) V, (Cassiod.): apostoli exemplo H. 

prosunt ABKGV Cassiod.®*^- : prosint H, Cassiod.''"'^ (recte). 

injideli ABKG Sedul.: alia VH Cassiod.^*'- (= vg): aliena 

In the case of 1 Cor. x 27 the omission of ad cenam is supported 
by AK V Cassiod. (= vg), while ad cenam is present in BH. This is 
a somewhat difficult case. K agrees with B in amplius aliquid for 
aliquid amplius of the others. 

But if these examples show that the Freiburg fragments keep 
good company, the text is nevertheless characterized by considerable 
errors: abstinere repeated after salutis (viiii 15), nee adnuntiem after 
adnuntiem (viiii 18), quod after licet (x 24), esse after conscien- 
tiam (x 25). The following omissions occur : legem after qui 
(viiii 20), ut aiter fratres (x 24). These readings are wrong: in 
for 7ne (viiii 15), fecisse fovfecistis (viiii 16), immolare for immolate 
(x 25). 


If we compare the Freiburg text with A and B in some detail, 
weshallfind that in fifteen cases it disagrees with their joint evidence: 
in at least fourteen of the cases, it is undoubtedly wrong \ But there 
are three and a half cases where K agrees with B against A. They 
are amplius aliquid (1 Cor. viiii 16), eriim omitted (1 Cor. x 26 = vg), 
ergo (1 Cor. x 30), parte cipo (1 Cor. x 30 = vg participo) where B 
has participor. 

In these cases the full evidence is: 

amplius aliquid BKG: aliquid amplius AVH. 

enim omitted BKVH Cassiod. (=vg): enim AG Sedul. 

ergo BKGM, N {corr.) R {corr.) Sedul.^''- : ego AVHC Cassiod. 

parte cipo {participo) KVH Cassiod. Sedul. (= vg): percipio A: 
participor BG. 

In the first case, K is probably right, as it is the lectio dilfficilior, 
and we can see why the alteration was made in the others. In the 
second it is natural to regard the non-Vulgate reading as right, 
especially as the Psalm itself also lacks the enim, and enim is a 
word which is apt to be omitted : also 70/9 appears to be everywhere 
present in the Greek. The next case is rather more difficult to judge. 
No doubt ergo is an incorrect reading, whether it be an anticipation 
of the ergo of verse 31 or a mere palaeographical mistake, as 
Tischendorf conjectured. But it is not easy to say whether the 
reading is Pelagian or not. Sound criticism will, in the fourth case, 
I think, hold to participor: percipio is one of those aberrations of 
A to which reference has been already made. In only one of the 
four places, therefore, is K with absolute certainty right. There 
are no places where K agrees with A against B. 

It is a misfortune that so little of this manuscript survives; also 
it is to be regretted that what does survive should belong to a part 
where Pseudo-Jerome has made no interpolations. It is really im- 
possible, therefore, to deny that the complete Freiburg codex was 
a Pseudo-Jerome, but on the whole the evidence, I think, ftivours 
the view that we have here the scanty relics of another Pelagius 
of the original extent, not very closely related to any MS of which 
I have knowledge. 

1 The possible exception is aliut quid (1 Cor. x 31) for aliquid. 


(b) Manuscripts of Interpolated Forms 
(1) No. 73 in the Stiftshibliothek at St Gall (G) . 

The manuscript is thus described by the late Dr G. Scherrer in 
the Verzeichniss der Handschriften der Stiftshibliothek von St Gallen 
(Halle, 1875) p. 31: 

'73. Pgm. 2°. s. IX incip.: 262 pages\ double columns, with 
corrections. Glossae incerti auctoris in epistolas sancti Pauli 
(Incip. pag. 3: Paulus. expo. Querimus quare Paulus scribat etc.).' 

To this it may be added that it measures 35 x 25 cm., that it 
was written by two scribes, and that there are commonly 37 lines 
to the page, though on page 13 there are as many as 47. The real 
beginning of the MS on page 1 is iNCiP/r omnium epistolarum. 
Primum queritur quare. . .discipuli apostolontm, page 2 being blank. 

The first sheet contains now only six leaves (twelve pages) 

13 5 

I I I 

7 9 11 

I I 1 

the second is a normal quaternion, lettered B at the foot of the 
second column: so with quaternions C, D, E, F, G, H. The next 
sheet I is a trinio signed at the end (foot of p. 136): K and L 
are normal and signed at the end. M takes this form: 

170 172 174 176 

178 180 182 

That is, the last leaf has been cut out, but a different scribe has 
signed the second last leaf M: N, O, P are normal. The end of P 
coincides with the end of the exposition of Philemon (p. 230 a): 
p. 230 b is blank, and the exposition of Hebrews commences at the 
top of the new quaternion Q (p. 231 a). The last quaternion of all 
is unsigned. 

The initial words are as described above. The closing parts are: 

230 a crescit quoties loquitur et auditur. FINIT EPISTOLA 
pauli ad philimonem. 

231 a INCIPIT argumentum ad hebreos. In primis dicendum 

^ The St Gall MSS, like those of Ghent, are numbered by pages, not folia. 


est cur. ..260 a baptismi, pgnitentiae et perseuerantiae. finit ex- 


poetical matter. Page 262 is blank. 

Titles on the cover are Appostolus paulus glosatus and Epistole 
B. Pauli glossatae. Strong wooden boards, covered with leather. 

Besides the modern signature 73 on the outer cover, there are 
to be found (fol. 1) the bookplate with Sig. Monaste. sane, galli and 
also the old shelfmark D. n. 23, under which the new one has been 
added in red letters. On page 2, which is otherwise blank, a fifteenth 
century hand has written Apostolus patdus glosatus n 4. 

There can be little doubt that Zimmer is right in identifying 
this manuscript with that referred to in the old St Gall catalogue 
of about the year 850, preserved in manuscript 728 of the St Gall 
library, as 

Expositio pelagii super omnes epistolas pauli in uolumine D. 
But Zimmer makes a curious mistake about this entry. He states 
that it was added to the orfginal catalogue along with two items 
Tractatus Origenis in Genesim, Exodiun et Leuiticum in uolumine I 
and Item, Tractatus Origenis super epistolam ad Romanos uolumen 
optimum. As a matter of fact, personal examination made it clear 
to me that, while the second Origen entry is certainly later, the 
Pelagius entry is part of the original catalogue 2. Zimmer's argument 
that the Pelagius manuscript was incorporated in the library between 
850 and 872 thus falls to the ground, and there is nothing to hinder 
us from assigning it to the first half of the ninth century, to which 
palaeographically it belongs. 

The manuscript is written in a continental hand, by more than 
one scribe. As Zimmer has pointed out^ one scribe wrote quaternions 
A — I), N — P, and another E — M, Q — R; or, in other words, the 
first wrote the expositions of Romans, Colossians, 1, 2 Thessalonians, 
1, 2 Timothy, Titus, Philemon, while the other copied the exposi- 

1 The catalogue published in G. Becker, Catalogi Bibliothecarum Antiqui no. 22 
pp. 43 ff., P. Lehmann, Mittelalterliche Bihliothekskatalofje Deutschlands nnd der 
Schweiz... I Bd (Miinchen, 1918) p. 76 1. 6. 

^ I am now confirmed by Lehmann, loc. cit. Zimmer's error, derived from 
Becker, is repeated by Koetschau, Origenes Werke Bd v (Leipzig, 1913) p. lii, whose 
discussion is vitiated by another error of Becker's, the attribution of catalogue no. 15 
to St Gall, whereas it belongs to Reichenau (see Die Reichenauer Handscliriften 
Bd III (Leipzig, 1916) pp. 97—103). 

3 Pp. 238 fl". 




tions of 1, 2 Corinthians, Galatians, Ephesians, Philippians, Hebrews. 
No doubt these two scribes worked concurrently, in order to get 
their task finished as quickly as possible, and there is no reason to 
suppose that the scriptorium where the manuscript was executed, 
was any other than the St Gall scriptorium itself. A third hand 
writes occasional passages, e.g. p. 40 a, 11. 21 — 25, Radicis patrum 
to per te stant. A fourth hand, perhaps to be identified with the 
corrector of the codex wrote pp. 55 b, 1. 1 to 56 a, 1. 3 (end of Romans). 
The orthography of the manuscript as well as the abbreviations 
show many Irish ^ symptoms such as we expect to find at St Gall. 
We shall consider the orthography immediately in an endeavour 
to trace the ancestry of the manuscript. The abbreviations employed 
are these: 

apostolus apos (dat. ) (abl. p. 223 b), apl 
(coiT. ) (S-L2), apols, apis (S-L), aplsts 
(corrector), aplsls (p. 47 b) 

apud ap 

aut a (ouce) (S-L) 

autem au nearly always (S-L), aut 
very rare (S-L), \r also very rai-e, 
and only above the line in additions 

capitulum 1c 

carissimi criui, cmi (corrector kiTii), 
kmi (SJ.) 

cetera c& 

Christiis xps 

cmn c 

Dauid d9 (S-L) 

de d 

deus ds (S-L) 

dicimus dms 

dicit die (S-L), dit (S-L), dt 

dicitur dr (S-L) 

diciint dnt 

dicuntur diir 

dominns dns (S-L) 

ecclesia ecla 

eius ?) (once only, s.l.) (S-L) 

enim ff (S-L) 

lepis (accus, sing.), epos 

episcopus < (accus. plur.) 
(eps etc. 

epistula epTs, epla, epl 

ergo g 

esse ee (S-L) 

est e (S-L), ^ (S-L) 

et 7 (corrected at least once to et) 

explicit expli etc. 

expositio exp, expos^expo 

frater frt (S-L), ff 'Jratre,' frm 
'fratreni,' ffs (S-L), ff, fras (rare) 
(S-L), fs, frf all for 'fratres.' 

haec h (altered once by a cori'ector 
to hec) (S-L) 

hoc li, H 

hoc est ho -=- (S-L) 

lesus ihs 

Johannes iofi, ih 

Israhel _ i.sri (S-L) 

item it (very characteristic) 

nieus ms (S-L) 

mihi m 

misericordia mla (S-L) 
nomen noe, 'nomine' 
noster nni, nfni 'nostrum' etc. 

(S-L both) 
icester ur, 'uester,' ura etc. (S-L) 
7iicnc nc 

omnipotens omps (S-L) 
omnis oms (S-L) (p. 160b, corrected 
by corrector to omis (S-L)), oml(?); 
om (S-L), oms (S-L) 'omnes'; oma 
(S-L), omia (S-L) 'omnia' 

1 See also Zimmer, pp. 232 ff. 

2 The letters S-L are added to those abbreviations which are definitely attested for 
St Gall MSS in Fr. Steft'eus, 'Die Abkiirzungen in den lateinischen Handschriften 
des 8. und9. Jahrhunderts in St. Gallon' (Zentralblatt filr BibUothekswesen, Bd xxx 
[1913] pp. 477—488) ; W. M. Lindsay, ' Note on the Preceding Article' (ibid. pp. 488— 
490). These scholars did not use MS 73. 




Paulus pan 

per p (S-L) 

post p' (S-L) 

prae p (S-L) 

presbiter pfbi 'presbiteri' 

propter pt (twice), pp (S-L) (once, 
altered by corrector to ppt (S-L)) 

qiuie q: q- q- (first and second S-L) 

quam numq. ' numquam ' 

qv,ando qn (expanded once by cor- 
rector) (S-L) 

que q: (S-L) q- (S-L) 

qiiem q. (several times) (S-L) 

qui q (S-L) 

quia ^, q: (S-L), qi 

qxmd qd (S-L) 

quoniam qiTi (S-L), quo (very rare) 

(S-L), qniu (rare) (S-L), qm (once) 
reliqua rq, rel, req, rl, reliq, rlq, relq 
saeculum scla 'saecula' (S-L), saclo 

' saeculo ' 

Syllable Symbols : 

hunt Bt, bt (S-L), bn+ 

con c (corrector only, except p. 142 a) 

e d 'de' 

en m 'men' (S-L) 
er B 'ber' (S-L), t 'ter' (S-L), u 'uer' 

is b 'bis' (S-L) 
it die 'dicit' (S-L) 
ill suprascript stroke (S-L) 
n suprascript stroke (S-L) 
OS see MS 

sanctus scs, etc. 

secundum seed (S-L), scdm (S-L), 

sed S' (very characteristic) (Swiss, 

perhaps from Italy) 
sicut sic (S-L) 

significat sig, sigii (both end of line) 
spiritaliter spitaliter 
spiritus sps, spiis 'spii'itus' (genit.) 

(S-L), spurn 'spiritum' 
subauditur sub 
sunt s, st (S-L both) 
tamen tfi (S-L) 
tantum tm 

tempore tep 

tunc tc (S-L) 

icel 1 (very characteristic) (S-L) 

uero u 

versus uefs, uf ' 

ri suprascript i ; p 'pri' (S-L) 

rum if (S-L) 

runt f (S-L), rt (S-L) 

ur rn2 'mur,' t' (S-L) 'tur,' f^ (second 

scribe) (S-L) 
us cui' 'cuius' 

b: (S-L), b' 'bus'; m' (S-L), m, 

m; 'mus' 
11+ 'nus' (S-L) 

also for OS : p'sidebit ' possidebit,' 
p'tea 'postea,' etc. (S-L) 

The instances to which the letters (S-L) are added are sufficient 
to show that MS 73 is a product of the St Gall scriptorium. Certain 
of the others have been taken over inadvertently from the archetype, 
and may give us indications of the ancestry of our manuscript. 
There can be little doubt that the symbols for ajjud, autem (third), 
cum, dicimiis, dicit (third), dicunt, dicuntur, ergo, et, hoc, mihi, nomen, 
nunc, quam, quia (third), secundum (third), tantum, uero, employed 
in this manuscript, were copied from an immediate ancestor in Irish 
script, which lay before the scribe. But I think traces of two earlier 
copies also show themselves. The symbols for apostolus (last two), 
de, ecclesia, item- on the whole favour the idea of a Visigothic strain. 

1 Wrongly interpreted by Zimmer (p. 382) as 'urbe.' 

2 See Notae Latinae, p. 115 for another St Gall example. The standard works 
on abbreviation are this and the earlier work of Traube, Nomina Sacra (Miinchen, 


What I conceive to have been the case is that the immediate parent 
of the Irish MS was Visigothic, and there is nothing antecedently 
improbable in this view. The connexion between Ireland and Spain 
was very considerable. But we can go I think even behind this 
Visigothic manuscript which probably belonged to the period 600 
to 800. Certain of the abbreviations take us a stage farther back 
still; I refer to such as those for apostolus (the first), episcopus (the 
first), epistula (the first), jji^eshiteri. In the first, second and 
third cases the abbreviation by suspension recalls a very early 
stage of transmission, perhaps a half-uncial of the sixth century. 
Farther back than this we cannot go. These arguments drawn 
fi'om the abbreviations employed can be powerfully reinforced 
from the orthography and from corruptions present in the 
St Gall MS. 

Zimmer has already collected instances from the manuscript of 
what he considers Irish orthography \ I do not regard all of them 
as equally cogent, and have therefore studied the matter indepen- 
dently. But some of them at any rate will stand. Confusion of 
vowels is one of the most striking characteristics of Irish manu- 
scripts, and of this we find plentiful illustration. The. most common 
confusions, abundantly exampled in this manuscript, are a for e, 
e for a, i for e, e for i; less common, but yet frequent are o for u, 
u for 0, a for o, y for i; others, which need not be specially Irish, 
and are still less frequent, are o for a, u for a, a for u, e for o, o for e, 
e for u, u for e, u for i, i for y. Another characteristic of Irish MSS 
is the doubling of single consonants, and the 'singling' of double 
consonants. Of these phenomena the following examples occur 
among many others: (a) ahssens, circumcissus, cassibus, conmissistis, 
pertulli, iddolis; (b) posunt, corumpit, nulius, comutare, comouere, 
mitatur. The following spellings may be safely attributed to the 
Irish exemplar: plasphemus, ponis (for bonis) (p. 135 b), spahnus 
{for psalmus), tetinere (for detinere), contempnare (frequent for con- 
demnare): possibly also redient (for redigeat), neglientia, a.nd ewiere 
(for exigerey. 

^ Pp. 234 f. I propose to use some of them to indicate Visigothic ancestry. My 
own lists are quite independent of his. 

- This last type of spelling is very common in C of the Pseudo-Augustinian 
Quaestiones, which appears to be a copy of an L'ish exemplar (SB. der Kais. Akad. 
der Wiss. in Wien, Bd cxlix [1905] (I) p. 9). 


That an Irish original was in front of the scribes of St Gall 73 
is also amply proved by certain misreadings of the exemplai*. Only in 
reading an Irish (Insular) script could the following confusions arise: 

n for r : con for cor (p. 80 b) ; condelectante for corde laetante 
(p. 94 a); ma for tVa (p. 162 a). 

r for 71 : deperderent for dependerent (p. 45 b) ; liherter for li- 
benter (p. 170 a); oppiniorum for opinionum (p. 229 b). 

ri for n: tripherium for triphenam (p. 54b), etc.; in for ri: 
comparam for comparari (p. 142 b); rwi for m: erminentem for 
eminentem (p. 177 a). 

^ for 3: haptigentur (p. 85 b), haptigantLir (p. 102 b), eaange- 
ligahant (p. 171 b). 

3for^: euaa^eliwn (p. 70h), euari}eli^et (p. 137 b), eiiai^eliyvt 
(p. 139 a), euani,eli}aui (p. 144 b). 

There are also errors which could arise only fi'om a neglect or 
misunderstanding of Irish abbreviation symbols. Such are in for 
autem (hf) (p. 163 b); the frequent omission of autein (hr) and enim 
(tt), and the substitution of one for the other; the omission of 
eius (^) (p. 226b); the use of quod (q.) for quia (g.) (p. 174 a); 
quam (q) for quod (q.) (pp. 52a, 228a); uel (I) for id est (i-) 
(p. 131 b); pj^ae (p) former (jf) (pp. 105b, 150 a)\ 

That Visigothic influence lies behind the Irish influence is 
suggested by the following symptoms: bebet for bibit (p. 87 a), 
pohidus for poptdus (p. 86 b), suberbia; ocidtus (regular in this MS); 
abdicaberint; accipiad,uenissed ; facillantes iov uacillantes (p. 160b), 
defortium for diuortium (p. 165b)-; simalagrorum (p. d23i),fugata 
for fucata (p. 170 b); the wrong presence of /i before a vowel, 
especially at the beginning of a word, as in haccubitas, hapud, 
haudeo, helimosina {helymosina), herudieris, hodium, homnis, horans, 
hostendo, husque, adholatio, exhistimo, as well as its absence where 
it ought to be present, as in abere, eves, omo, ova, imianitas, sub- 
traebat, distraere^; quohabitatio for cohabitatio, etc.; aceruitate, 

1 quo for quo ( = quoniam), p. 4 a, is against alike an Irish and a Visigothic 
exemplar. Probably our scribe has simply forgotten to write the cross-stroke. 

2 These instances are valuable as illustrations of Prof. Lindsay's deuoret for 
deforet of the Visigothic and other MSS of a fragment of Lucilius in Isidore, Etym. 
XIX 4 § 10 (cf. Classical Quarterly vol. v (1911) p. 97). 

3 I am well aware that these phenomena are not confined to Visigothic MSS, but 
that they are specially characteristic of these, there seems to be no doubt ; see an 
excellent example in A. E. Burn's Niceta of Remcsiana (Cambr. 1905) p. Ixxxviiin. 2. 


adorauit, seriiauimus, separaidt, liuertas (p. 88 a) on the one hand, 
with hrehi for breui, donahit, seruahit, dehitiim for diuitum (p. 64 b), 
on the others The confusion between semi-consonantal i and g is 
also to be traced to Visigothic influence'-: proienie for progeine, 
iuium for iugum on the one hand, and gam for iam, geiiinis for 
ieiuniis (p. 85 b), agunt for aiiint, geiuniis (p. 132 b), agit for ait 
(p. 176 a), if agit be not the true text, as it is on p. 186 a. It is 
perhaps hardly safe to call cartelarium on p. 223 a {=cartularium) 
a Spanish symptom, but the spelling is not otherwise recorded. 
Finally, per and pro could only be corrupted from a Visigothic 
exemplar, for it is in Yisigothic alone that the s^'mbol which in- 
dicates pro elsewhere, has the value o^ per; we find per for pro on 
p. 195 b, and pro for per on pp. 120 a, 153 a, 229 b. 

It is strange to find any distinct characteristics of the very 
early period, after the blighting influence of a Visigothic and an 
Irish scribe has worked its will on a text; yet it appears that a few 
such traces remain, in the following venerable spellings, which 
must have passed unscathed through the various stages: Danihelo 
(once or twice), Danielmn; Eleazarus (for the ordinary Lazarus of 
Luke xvi); Isac (nearly always); prode est (pp. 95 a, 102 b, 103 a); 
unianimis (always); iiul (for uidt) (pp. 81 b, 127 b). The numerous 
confusions between r and s may have occurred either at this stage 
or at the Irish stage. 

I venture to think, then, that a study of the abbreviations and 
orthography of our manuscript shows that it was copied from an 
Irish exemplar, which in its turn was a copy of a Visigothic exem- 
plar, and that this Visigothic exemplar was copied from an early 
manuscript, say a semi-uncial of the sixth century. Whether the 
whole text as we have it was contained in this very old copy, is 
a question somewhat hard to answer. It is quite possible a priori 
that certain accretions took place at the Visigothic or at the Irish 
stage or at both. 

Some idea of the contents of this manuscript has already been 
given in the second chapter of this book. The codex contains 

^ Zimraer (pp. 234 f.) classifies these last as Irish spellings, perhaps rightly; but 
both views may be correct. 

2 See E. A. Lowe in SB. K. Bay. Akad. Wiss. 1910 (1'2) pp. 14 ff.; H. W. Garrod, 
Classical Review xxxv (1921) p. 40. 


nearly the whole of the real Pelagius, but in addition it contains 
most of the interpolations we have learned to associate with the 
Pseudo-Jerome form, particularly with this form as it appears in 
the longer branch of that family (H,). It contains, however, on the 
one hand more, and on the other hand less than Pseudo- Jerome. 
As Zimmer has observed, it generally omits, in the exposition of 
First Corinthians, one of the two explanations given by Pseudo- 
Jerome ^ In the case of that Epistle it corresponds almost exactly 
to the manuscripts of the uninterpolated form with one exception. 
It interpolates the chapter headings generally found in Vulgate 
manuscripts-. On the other hand, from the EjDistle to the Ephesians 
onwards, it is interpolated from the genuine commentaries of 
Jerome on that Epistle, the Epistle to Titus and the Epistle to 
Philemon^ It is interesting to observe that Jerome's commentary 
on Galatians was not used. This fact suggests that the inter- 
polator had only the other three at his disposal. In this connexion 
I ought to point out that several MSS of Jerome in Eph., Tit., 
Philem. together, without Gal., are still in existence. They are 
these: Karlsruhe, Codex Augiensis Lxxxi (saec. ix in.); St Gall 
129 (saec. ix); Kbln LViii (Darmst. 2052) (saec. ix); Wolfenbuttel 
13 Weissenb. (saec. x); Florence Laur. plut. xviii dext. cod. ix 
(saec. xiii)*. What inference can we draw ? This, I think. As all 
the places to which the earlier manuscripts belong are within the 
zone of Irish influence, it was the compiler of the Irish exemplar 
of our manuscript who extracted them from a manuscript of Hier. 
in Eph., Tit., Philem. in his possession, and added them to what 
he found in the Visigothic codex. In other words these inter- 
polations were absent at the Visigothic and the earlier stages. In 
addition to portions of genuine Jerome ^ our manuscript contains 
extracts from Augustine'' and Gregory I It is probable that these 
emanate from the same Irish interpolator as added the passages 

^ Zimmer, pp. 246 f. - See Zimmer, pp. 249 f. 

The passages are indicated in Zimmer's footnotes from p. 357 onwards. 

■* I made a list for my projected Vienna edition. 

^ To which I must add one from epist. 28 g§ 4, 5, unidentified by Zimmer, on 
p. 229 b, which is shared with the longer Pseudo-Jerome form. 

6 'ut augustinus {ex agustinus) dicit,' p. 165 a; ' augustinus dicit,' p. 201a 
( = in 2 Thess. ii 8), 205 b, 'beatus augustinus dicit,' p. 207 b. 

" 'gregorio diceute,' p. 222 b. 


from genuine Jerome. If this be so, the date of the Irish MS must 
be post- Gregorian, and there can be little doubt that it belongs to 
a date not earlier than the middle of the seventh century. The 
Jerome quotations are added without the name of Jerome\ a fact 
which suggests that the Irish scribe found the whole work under 
Jerome's name, and therefore did not add the name to extra notes 
derived from other works by the same author. If this be so, then 
our manuscript is a (modified) member of the second family of 
Pseudo-Jerome MSS (H2), to which it is otherwise closely related, 
as Hellmann contended against Zimmer". 

The St Gall MS is like the second family of Pseudo-Jerome 
MSS in two other respects ; it gives the Epistles in the normal 
order, and it contains an (un-pelagian) commentary on the Epistle 
to the Hebrews, whereas the first family contains none. This 
commentary is not, however, identical in both cases. The St Gall 
commentary cites Cyprian, Augustine, Gregory of Nazianzus, and 
Gregory the Great, and is therefore not older than 650. Zimmer 
is no doubt right in regarding it as an Irish production^ The 
St Gall commentary on Hebrews is also preserved by itself in 
a manuscript at Wolfenbtittel, as Riggenbach was the first to 
point out. This is the MS already mentioned, 4097, Weissenb. 13 
(saec. x), but the part already referred to is really quite a dif- 
ferent MS that has been bound up with the commentary on 
Hebrews*. Riggenbach has also proved that, though not identical, 
the Hebrews commentary in the St Gall MS is very closely related 
to that in the second family of Pseudo- Jerome MSS^ and, further, 
he has done a great service to our studies by showing that the 
Pseudo-Jerome form is often primary and the St Gall form 
secondary. His conclusion is that they are both revisions of one 
and the same original, and he points out that in Pseudo-Jerome 

1 With one exception, p. 201 a, not from a commentary ad loc. : ' secundum ihero- 
nimum dicentem.' 

■^ Sedulius Scottits (Miinchen, 1906) pp. 153 ff. Hellmann, however, knew only 
the shorter form of Pseudo-Jerome (Hj), as published, which makes his discovery 
all the more acute. 

•* Zimmer, p. 276. 

■• Eiggenbach, Die liltesten lateinischen Kommentare zum H ebrlierbrie f { = Z&hn^s 
Forschungen zur Gescli. des nt. Kanons, viii Teil) (Leipzig, 1907) pp. 12 f. 

5 Op. cit. pp. 202 ff. 


no name of a Father occurs. Later in the same book he shows 
a close relationship between Sedulius Scottus's compilation on 
Hebrews and that of the St Gall MS. 

The textual character of the St Gall MS has already been 
illustrated by Zimmer and Hellmann with the aid of such materials 
as were at their disposal. I propose now to study their lists^ in 
the light of the fresh materials now accessible. The general effect 
of the new evidence is to clarify the situation. We shall follow the 
ordinary maxim that community of error implies community of 

Further Agreements of G (aS^^ Gall 73) with H (Pseudo- Jerome) 
where both are ivrong : 
Rom. i 3 addendum GH*V Sedul. : addendo^. 
exclusit G ; excUidit H : extinxit. 
recedentes GH Cassiod. : recedens. 
adhiheatur GH : id habeatur, 
estis GH : eritis. 

libertatein Christianam GH* : libertate Christiana, 
legi GH Sedul. : lege. 
22 salutem G (salute Hj saluti Hg) infii^mi : infirmi 

ostenditur GH : uidetur. 
projiciatis GH : profetetis. 
delicti GH : uindicandum. 
Item cognitiones — 
— destruendas dicit GH : om.* 
Eph. iii 18 diligant GH : eligant. 
2 Tim. iiii 15 resistit GH : restitit. 
Agreements of G luith Hi tuhere both are wrong : 

Rom. i 4 est GHj : portenditur {pertendit\_ur^ Hg). 
vii 8 obliuione[m] erat GHj : obliuionem ierat. 
viii 6 parat GHj* Cassiod. : parit. 

xvi 24 commoneret GHj : commemoraret {commemoretH.^. 
1 Thess. iiii 6 fratri GHj : pari. 

1 Zimmer, pp. 230ff., 243 £f.; Hellmann, pp. 153 ff. 
- The evidence for the true readings will be given in vol. ii ad locos. 
^ The examples of that type of error which consists in interpolation are very 

s. p. 16 


iiii 1 

vi 14 

xiii 1 

xiiii 4 


XV 21 

1 Cor. 

xiiii 1 

2 Cor. 

X 1 



Agreements o/G with Hg where both are wrong : 
2 Cor. iii 5 se nihil GH., : nihil se. 

rusticani GH2R : rustici. 
Gal. V 18 lea; nobis GH., : nobis lex. 

Phil, i 20 et uita nostra — pertinet Christi om. GHo : le- 

22 si uinere — operandi oin. GH2 : legendiim. 
ii 22 ut qiialis sit no- 

ueritis om. GH2 : legendum. 

23 mode enim in- 

certus sum om. GH., : legendum. 

From all this evidence it is clear that the St Gall MS is con- 
siderably different from Pseudo-Jerome. Before we pass on to 
compare it with other MSS, we can dispose of other differences 
not already mentioned. It gives the famous note on Romans v 15 
Plus praeualuit etc., which is wanting from every Pseudo-Jerome 
manuscripts It has the Pelagian Pi^imum quaeritur prologue, 
somewhat mutilated, and the Pelagian prologues to First Timothy 
and Titus. It has the Marcionite prologues to First Corinthians, 
Galatians, First Thessalonians and Philemon, and the Pseudo- 
Marcionite (catholic) prologues to Second Thessalonians and 
Second Timothy. Of the remaining epistles, Second Corinthians 
has the Pseudo-Marcion plus the Balliol prologue with some 
differences, Ephesians has the Marcionite plus a portion of Jerome, 
Philippians has the Marcionite plus the Pelagian prologue, Colos- 
sians has no preface, Hebrews has the usual Vulgate preface found 
also in Hj. From this it is clear that the basis of our manuscript 
was a copy of the Epistles or of Pseudo-Jerome, fitted with the 
usual Marcionite or Pseudo-Marcionite prologues. 

This last suggestion is supported by a partial stichometry in- 
corporated with the arguments. 

^ ^ fDCCCLXX^ 

2 Cor. i 


1 Here and elsewhere of course I except the Balliol MS, which is in a sen?e a 
Pseudo-Jerome. It is convenient to treat it by itself, as it differs so greatly from the 
longer manuscripts. 

- Probably this has become displaced and really refers to 1 Cor. It is given at 
the beginning of 2 Cor., whereas the other is given at the end of 2 Cor. 




2 Thess. 


1 Tim. 


2 Tim. 




It is obvious that this stichometry is in part corrupt, even if there 
are not traces of two systems here. There is some relationship 
between it and that which is given by the second family of Pseudo- 
Jerome manuscripts-, though only in two cases, Phil, (b) and 1 Tim., 
are the numbers absolutely identical. Both omit stichometry in 
five cases, only eight numbers being given in each case, and on 
the whole the same epistles are alluded to in both. It would be 
of the utmost importance to fix the date and place of origin of the 
stichometrical system here employed, but in the present state of 
our knowledg-e I am unable to do so. Meantime we must return 
to the consideration of the textual relations proper of our manu- 

Agreements of G and the Wilrzhurg glosses^ ivhere both are wrong : 

Rom. i 21 imaginem GWb : magnitudinem. 
2 Cor, X 16 ad alios GWb; alibi Cassiod.*^'- {om. Cassiod.°°''): 
The two also share various interpolations, e.g. at Eph. v 14*. 

Agreements o/G and the Vienna glosses^ where both are ivrong : 

1 Tim. vi 2 the interpolation in baptisnio GWn : om. 
Philem. 25 a considerable interpolation GWn : om. 

Agreement of Q and Cassiodorus where both are lurong : 

Rom. viiii 10 nati G Cassiod. : generati. 
The question of the Biblical text employed in the St Gall 
manuscript must now be considered. It is not pure Vulgate, but 
it approximates much more closely to the Vulgate than does 

' The first is given at the beginning, the second at the end. 
- See below, p. 270. ^ See below, pp. 326 ff. 

* Zimmer, pp. 262 ff., 270 f.: cf. also his notes on the Hebrews commentary 
(p. 275). 

3 See below, pp. 328 f . 



either A or B. Let us examine, for instance, Romans i and v 12 — 
21 with the aid of Wordsworth and White's apparatus. The 
following differences from the Vulgate occur in these sections : 
Rom. i 10 semper om. with B. 

11 ~ uohis gratiae BD^ etc. 
13 etiim vg. codd. A. 

17 aiiteni. 

18 et om. 

ueritatem del BD etc. 
iniustitia dei. 

20 intellectu. 

~ uirtas eius D etc. 
et ABD etc. 

21 qui with d*. 

23 incorruptihilis om. 

24 eficiant. 

25 mendacimn AD etc. 
avien om. AB etc. 

27 in (alt.) om. ABD etc. 

28 ea quae vg. codd. etc. 

29 repletos] + enim. 
iniquitate] + et B. 

fornicatione] + et B. 
32 solimi] + qui ABD etc. 

et] + qui ABD etc. (D however om. et). 
V 13 hoc imundo BD etc. 

15 gratia (alt.) ABD etc. 

16 iudicium] + quidem BD etc. 

17 in (pr.) om. vg. codd. AB. 
uitam AD etc. 

18 in (pr.) eras. vg. codd. Sedul. 
et om. Aug. ter. 

in (tert.) eras. vg. codd. Sedul. 

19 oboedientiam BD Sedul. etc. 
constituiuntur (sic) : constituuntur vg. codd. etc. 

20 super habundabit". 

21 et om. 

^ For the meaning of these symbols see chap. iv. 
2 See p. 238, above. 


The variants unparalleled in other authorities may safely be re- 
garded as merely graphic, or inspired by such intelligence as the 
scribes possessed. It will not escape notice that among the 
remainder, twenty-five in number*, there are ten shared with A, 
fifteen shared with B, and thirteen with the Book of Armagh (D), 
which we have discovered to be in close relationship with the text 
used by Pelagius himself^ The St Gall text is then, no doubt, 
like others which will come before us, and like the text in the 
Reichenau MS (A), the original Pelagian text as considerably 
revised to bring it into harmony with the Vulgate. If the com- 
mentary in the St Gall MS was really built round a separate copy 
of the Epistles, then this copy may very well have been an Irish 
text, which would naturally show relationship with the Book of 
Armagh. It has been shown above that Gildas used such a text, 
and the evidence, alike that already given, and that which is yet 
to be provided, shows that Sedulius the Irishman's text was of the 
same kindred. But more probably our MS represents what was 
a good Pseudo-Jerome at the Visigothic stage, and the other 
comments were added in the margin of our 7th or 8th century 
Irish copy of it. (See further under Pseudo-Jerome.) 

(2) No. 653 in the Bibliotheque Nationale at Pai^is (V) 

The wonderful manuscript now numbered 653 in the Latin 
collection of the Bibliotheque Nationale, bears also three earlier 
shelf-marks, the earliest of all being dlxxxviii, the next in age 
628 (belonging to the year 1645), and the next 3939 (belonging 
to the year 1682)^ It reached the Bibliotheque Royale from 
Italy during the reign of Henri II (1547 — 1559), whose arms it 
bears on the elaborate binding of that date, to which the gilding 
of the leaves may also be assigned. It belongs to the end of the 
eighth or the beginning of the ninth century^ It now contains 
only 296 folia, though from errors in numeration it appears to 

1 See chap, iv, pp. 126 etc. 

^ I am indebted to Monsieur H. Omont's kindness for the dates of these shelf- 

* So Diimmler, Poetae Latini Aevi GaroUni (Mon. Germ. Hist.) torn, i pars prior 
(Berol. 1880) p. 89; and E. A. Lowe, Studio, Palaeographica (SB. Bay. Akad. Jhrg. 
1910 [12 Abh.]) p. 86; W. M. Lindsay, Notae Latinae (Cambr. 1915) p. 471, 'saec. 
VIII '; so also M. Mellot in Nouveaii Traite de Diplomatique t. iii (Paris, 1757) p. 65. 




contain fewer still, namely 292. The leaves measure 27 cm, by 
18 cm., the written part 224 mm. by 12 — 14 cm. 

The quaternions are regular up to and including XIIII (ending 

on fol. 112v), and are signed thus: MiilliMii in the middle of the 

foot of the last page of the quaternion. Then follows a quinion 
arranged thus : 

113 114 115 116 117 118 ; 119 120 121 122 x x 

This is signed I ; then there are regular quaternions signed II, III ; 
then five leaves together thus, and unsigned : 

189 140 141 I 142 143 x 

then regular quaternions signed iiiliiii-, iiiilliiiii, iiiillliiii , etc.. 

till we reach iiiiVIIIIiiii (ending on fol. 212 v); then a binion un- 
signed, and then an ordinary quaternion signed •!• at the right 
lower comer of f. 224 v; then : 

225 226 227 228 

229 230 231 232 233 

signed i H I at the right lower corner; then a set of regular quater- 


VII i, 


nions, i III I , j nil i , i V i , 

(f. 288 v); then a binion unsigned, 

X X 289 290 i 291 292 x x 


which finishes the codex, which is imperfect, two outer conjugate 
leaves having been lost from the last quaternion now represented ^ 
The number of extant leaves, then, may be computed thus : 
112 + 10 + 16 + 5 + 72 + 4 + 8 + 9 + 56 + 4 = 296. 

^ This loss is very old, probably older than the time of Henri II, and certainly 
prior to the saec. xv — xvi (?) foliation, which numbers fol. 288 by pp. 8 and fol. 289 
by pp. 9. The folia themselves were numbered by Arabic numbers, probably in the 
seventeenth century. 


There is a fol. 169 and a fol. 160 bis, a leaf is unnumbered between 
fol. 183 and fol. 184, another is unnumbered between fol. 189 and 
fol. 190, and yet another is unnumbered between fol. 287 and fol. 
288. We thus see how the numerator is four folia short with his 
292 ff. 

Fol. Ir contains a short poem addressed, it appears, to Charle- 
magne himself ^ and the authors of the Nouveau Traite make the 
probable conjecture that this volume was a present to that 
Emperor. He was certainly active in Italy in the encouragement 
of learning from A.D. 776 at least^. 

On fol. Iv there is the following table of contents (in uncials): 














Fol. 2r begins with the shelf-marks already referred to: DLXXXVill 
stroked out by the hand that wrote 628: Primum quaei^itur pro- 
logue — ending on the middle of fol. 3v, the rest of which is 
occupied with the Roniani sunt argument in uncials; fol. 4r — 
fol. 5r (middle) contain the Romani ex ludaeis prologue; fol. 5r 
another version of the Romani sunt; then a tractate beginning 
Verhum caro factum est and ending esse non desinit, and (f 6r) 

1 This poem was published in the Nouveau Traite t. in (Paris, 1757) p. 78, and 
by Diimmler in Poetae Latini Aevi Carciini torn, i p. 1 (Berol. 1880) p. 98. 

2 \v. Wattenbach, DeiUschlands Geschtchtsquellen im Mittelalter . . . i Bd 7 Aufl. 
(Stuttgart and Berlin, 1904) p. 168. He died 11 Jan. 802. 

3 This table disguises the fact that iu the body of the MS the epistles are in the 
Pelagian order : Phil. 1, 2 Thess. Col. 


another beginning De numero apostolorum and ending partis vieri- 
tum 7-epperitur (fol. 6v)\ Then, after a fourteenth century hand's 
insertion of the words ' Incipit expositio in epistola ad romanos ' 
the commentary proper begins. After the commentary on Romans 
comes the Marcionite prologue to First Corinthians, and then in 
succession the following: Expositio Argumenti : Gorinthus metro- 
polis est Achaiae; et idcirco quod Corinthis (sic) scrihit, Ackiuis 
omnibus scrihit ; commentary on First Corinthians ; Pseudo- 
Marcionite argument to Second Corinthians, followed by the 
prologue to that epistle printed in the Benedictine (Migne) 
edition of Ambrosiaster, and found also in the interpolated MSS of 
that author^ as well as in our B; commentary on Second Corin- 
thians ; Marcionite argument to Galatians ; Pelagian prologue to 
Galatians; commentary on Galatians; Marcionite argument to 
Ephesians ; Pelagian prologue to Ephesians ; commentary on 
Ephesians ; Marcionite argument to Philippians : Pelagian pro- 
logue to Philippians ; commentary on Philippians ; Marcionite 
argument to First Thessalonians ; Pelagian prologue to First 
Thessalonians ; commentary on First Thessalonians ; Pseudo- 
Marcionite ai'gument to Second Thessalonians ; Pelagian prologue 
to Second Thessalonians ; commentary on Second Thessalonians ; 
Marcionite argument to Colossians ; Pelagian prologue to Colos- 
sians ; commentary on Colossians ; Pelagian prologue to First 
Timothy ; commentary on First Timothy ; Pelagian prologue to 
Second Timothy ; commentary on Second Timothy ; Pelagian 
prologue to Titus ; commentary on Titus [down to c. ii v. 11 where, 
after crapula, the first two leaves of a new quaternion are gone : 
these contained the rest of the Titus commentary, the prefatory 
matter to Philemon, and the Philemon commentary down to 
gaudium enim (v. 7)]; commentary on Hebrews [lost after c. iv v. 3]. 
This bald enumeration gives little idea of the multiplicity of -the 
contents, which we shall have to consider later. 

The conjoint opinion of palaeographers who have seen this 
manuscript is that it was written in Italy. ' Ainsi il paroit venir 
d'ltalie' is the verdict of the authors of the Nouveau Traite 

1 These tractates were published by me in Proc. Brit. Acad. vol. ii pp. 435 f. 
^ See chap, ii p. 57. 


(accepted by Dummler)^ Dr E. A. Lowe calls it 'a north Italian 
MS of about 800 A.D.-,' and assigns it hesitatingly to a Verona 
scriptorium^: Lindsay's attitude is identical^. 

The signatures of the quaternions suggest that at least three 
scribes took part in writing the codex. As a matter of fact, I be- 
lieve there were four, and that they distributed the work thus : 

1st scribe: fol. Ir (the Charlemagne poem), fol. 6v — 114v. 

2nd scribe : fol. 1 v — 6 v (end of prefatory matter). 

3rd scribe: fol. 115r— 142v. 

4th scribe: fol. 148 r— end (296 v). 
The writing of the first scribe ' shows cursive traditions : it uses i-longa, 
£) (for soft ti), the ligatures of ri, st etc. Characteristic is the r with 
the shoulder extending over the following letter. The second scribe 
lacks i-longa, £) , ligatures of ri, st etc. and represents the more 
modern tendency•^' To this it may be added that the third scribe 
uses the i-longa, both as the initial letter of a word and as repre- 
senting the semi-vocalic i, also the £| (soft ti) symbol, the ligatures 
of st etc., and the r with the shoulder extending over the following 
letter : the fourth scribe shares all these characteristics''. 

The following abbreviations occur in the manuscript. Where 
they are already attested for a Veronese scriptorium, I have added 
the letter V in brackets ^ 

aliter al al al~ ali (ff. 71 r etc.) alt caput cap (end of line f. 1 1 7 v) 

(f. 146 v) alite (f. 103 r) Christianus xpianis ' Christianis,' 

apostolus aposlu'apostolum' (f.l77r) etc_ 

autem au (very frequent) (V) au (f. Christus etc. xps etc. 

116v) (aute f. 16 v) Colosenses 'colo.senss' (f. 264 v) 

1 References above. Nouveau Traite vol. in, opposite p. 65, gives a plate repre- 
senting the subscription at the end of 1 Cor. and the title at the beginning of 2 Cor. 
of fol. 143 r. The authors also had enough intelligence to realise that the MS, though 
anonymous, contains 'I'exposition ou les commentaires de Pelage sur les epitres de 
S. Paul.' 

- Stiidia Palaeofjraphica p. 5 n. He also publishes a facsimile of a portion of one 
page (plate 2), given him by Prof. W. M. Lindsay. ^ Op. cit. p. 43. 

* Notae Latinae p. 471. He thinks Paris B. N. 9451 (which I have not seen) is 
from the same scriptorium. ^ Lowe, op. cit. p. 86. 

^ The Trustees of the Revision Surphis Fund in Oxford very kindly defrayed the 
cost of over 400 rotographs of this MS. 

" 'Abbreviature nel Minuscolo Veronese ' by Antonio Spagnolo in Zentralblatt filr 
Bibliothekswesen xxvii Jhrg. (1910) pp. 531 — 548 (with W. M. Lindsay's note, 
pp. 549—552) as corrected by the author in xxviii Jhrg. (1911) pp. 259 — 261. 




Corinthii corint 'Corinthios' 
deitas ditate 'deitate' (f. 108 r) 
deus etc. ds etc. 

dominus etc. diis etc. : note dnm 
dnis dnos (of secular lords once or 
twice, e.g. ff. 46 v, 213 v) 
ecclesia aeccia (fF. 80 v, 101 r) 
episcopits epi 'episcopi,' epm 'epis- 
copum' (fl'. 287 r, i87v bis), epos 
' episcopo_s_' (f. 270 r bis) (V) 
epistula epis (ff. 264 v, 279 r, 287 r, 

289 v) 
est -=- (ff. 62 V, 103 v s.l, 142 v, 145 r 
etc.) (V) pot -j- (end of line, f. 153 r) 

et k _ 

explicit expt, exp, expl 

expositio expo (f. 291 r bis) 

fratres ffs ( V) ff (ff. 1 22 r, 1 29 r etc.) 

gloria gla (f. 118 r, and six other 

times) (V), gla (f. 290 v) 'gloriam' 
hie deest M (in'text f. 141 v) 
hie pone hp (in margin f. 141 v) 
id est id (f. 153 v, end of line) (V) 
lesus etc. ills etc. 

ifiu 'lesum' (f. 216 r) 
ihus (ff 164 r, 176 r, 
indpit incp, inept, inc (f. 289 v) 
Israhel iftl (V), ihlis 'israhelis' 
Israhelita ililita (f. 52 r etc.) 
Israheliticus ifilitico 'israhelitico' (f. 

201 v) 
item it (V) 
meus ins (fourth scribe only, but 

freqiiently) (V) 
mihi never abbreviated ^ 
misericordia mia 'misericordiam' (f. 
mae 'misericordiae' (f. 
261 v), ma 'miseri- 
cordia,' ma 'miseri- 
cordiam' (used only 
by fourth scribe, and 
sometimes expanded 
by another hand) (V) 
\nobis nob (V) 
[uobis uob (V) 
non n (V) 

fnoster often written in full, with 
other divine names con- 
tracted, thus : dni nostri 
ihCi xpi 
n 'nostro' (f. 203 v) 
ner 'no-ster' (ff. 66 r, 109r) 

noris 'nostris' (f. 178r), 
nore 'nostrae' (f. 145 v), 
non 'nostri' (f. 229 v) 
nsr (five times, fourth 

scribe) (Spanish) 
ni 'nostri' (f. 221 r etc.), 
nili 'nostrum' (f. 230 r, 
and often), na 'nostra' 
(f. 149 V etc.); this type 
used by fourth scribe 
only, except that it occurs 
also f. 31 r 
nfi 'nostri,' 'nrae 'nostrae,' 
_etc. (V) _ 

uester ueri 'uestri' (once), uera 
'uestra' (f. 77 r only) 
ua 'uestra,' uis 'uestris' 
(each twice by fourth 
uri 'uestri,' ufo 'uestro,' 
V etcJV) 

omnipotens omptis 'omnipotentis' (f. 

269 v) (V) 

omnis oms 'omnes' (V), 'omnis' (ff. 

164 V, 165 r, 178 r, etc.) (V) 

oma 'omnia' (f. 278 v, and five 

times later) (V) 

paenitentia penitia ' paenitentiam ' 

(f. 175r) 
per ^_(V) 
p/'ae p (V) 
pro J) (V) 

propheta pph (f. 10 v, mg) 
propter prop (very frequently, be- 
ginning with f. 206 v) (V) 
ppr (f. 216 v) 
(pi'opt, end of line) 
(j)pt, end of line) (V) 
pter (ff. 118 r, 126 v) 
psalmus psl (once, f. 116 v) 
q^^e q: (V) 


g, (end of line, f. 209 r) (V) 

1 I mention this because mti is a characteristic Veronese abbreviation (Lindsay's 
note on Spagnolo [see p. 249, n. 7 above] p. 549). 

^ These appear to be the oldest known examples of this Italian abbreviation (see 
Lowe, Beneventan Script [Oxford, 1914] p. 208). 




qui q. (m 2 mg) (V) 
quod qd (once) (V) 
quoniam qm (V) 

qnm (f. 133 r) (V) 
saecidum saeli 'saeculi' (f. 142 v) 

sclra (V), scli 'saeculi' (f. 

290 r) (V), scla 'saecula' 

(V), sclru (f. 286 v only) 

sanctificatio scificationis 'sanctifica- 

tionis,' etc. 
sanctifico scificaret ' sanctificaret ' 
scificatis 'sanctificatis' 

sanctitas scitatem 'sanctitatem,' sci- 

tatis 'sanctitatis' 
sanctus etc. scs etc. 

sew ' sanctorum ' (f. 

247 v) 

Syllable Symbols : 

ae §(V) 

er t 'ter' (V) 

IS B 'bis' (f. 61 V, 70 V, 71 v, 112 r 

etc. etc.) (V) 
it dilig" 'diligit' (f. 127r) (V) 
m suprascript stroke (V) 
n suprascript stroke (V) 
u the V form superposed after q, as 

also in tialf-uncial script, e.g. qi 

(fol. 2v)i 
um rx 'rum' (V) 

secundum sedu ff. 158 r) 
spiritalis etc. spalis etc. 

spilis 'spiritalis' (f. 
169 r), spilia 'spiri- 
talia'(ff. 115v,129v) 
spitali ' spiritali ' (f. 
1 46 v), spltalem ' spi- 
ritalem' (f. 128 r), 
spiritaiiter spaliter 
spiritus etc. sps etc. 

spum 'spiritum' (f. 
123 v) 
sxint st (end of line once, f. 177 v, 
and twice in a correction, f. 114 v) 
uel ul (near end of line) (V) 

t 'tur' (flf. 44 r, 281 v) (very rare) 

tr 'tur' (end of flF. 158 r, 204 r, 
205 v) 



pS (f. 120v) eiJ (ff. 137 V, 194 r, 
197 r) b. (ff. 187 r, 200 r) mS 
(ff. 144 V, 153 r) t^ (flf. 177 r, 
203 r) IS (f. 209 r, end of line) 

The orthography of this manuscript is of great excellence. 
This does not mean that it is absolutely consistent throughout. 
We always find adh-, but on the contrary always ace- and arr-. 
Apart from these there is no consistency : we have adf- and aff-, 
adgr- and agn-, adl- (rare) and all-, adm- and am7n-, adn- and 
ann- {an-), adp- and app-, ads- and ass-, adt- and att-. We find 
conburo and conregno, but conl- and coll-, conm- and conim-, conp- 
and comp-, and corr- (always, except for conregno, conresuscito). 
There is hardly a trace of consistency in regard to either in- (nega- 
tive) or in- (preposition), even in the case of the same word. Thus 
we have inmanis and immanis, inmundus and immundus, inpune 
and impune, inperitus and itnperitus, but I believe always inl- ; 
and yet we find irritus, inipius, impie, impietas. The case is similar 
with the other in-, though there the unassimilated or uneuphonic 
forms greatly predominate. Impleo is regular, but inl- is even 

^ This is nqt, of course, a real abbreviation. I know no complete discussion of 
its use. 


invariable, while we have inpertio and impertio, inpedio and 
impedio, inpendo and impendo, inpugno and impugno, inplico and 
implico, inpono and impono, inhecillitas and imhecillitas ; inrideo, 
inrogo, but irritatio. Ohm- is (except once) invariable, but so is 
OCC-; yet we find ohf- and off-, ohp- (rare) and opp-, oht- (rare) and 
opt-. Per-, as far as used, is unassimilated, and the assimilated 
(euphonic) forms are almost invariable in compounds with suh-: 
thus we have sumministro, supplicium, suppono, supporto, supter, 
suptiliter, suptilitas, suscribo, while suhtilitas occurs, I believe, 
only once. Exs- is much less common than ex-, and the following 
inconsistencies occur: exsisto and existo, exspecto and expecto, 
exsurgo and exiirgo. The following forms are consistent : exsupero, 
exsequor on the one hand, execror, expolio, extasis, extinguo, exulto 
on the other. 

There are probably fewer traces of uneducated spelling in this 
manuscript than in any other with which we have to do. I have 
gathered together all the instances, and classify them here. They 
are, most of them, such as might be found in any MS of that date : 

aspirate omitted : exiheo, geennu, ebreus, aruspex, ospes, ospitor, ospiciian, 
istoria, istoricus, odiernus, ostia, ipocrisis, ortor, ymnus, retrao (twice). 

aspirate added : hahii, habundo, hahundanter, habundantia, herastus, honus, 

i for e: accipi, abolire, aduliscens (etc.), catheaiminus, debit (several times), 
discendo, distruo, dibilitas, discritio, dmis, dificio, diffirentia, dispero, deuie- 
risis, distituo, elimosijia, elimentum, erxibisco, famis, gentis, habit, hospis, 
herimus, innotisco, inrepraehinsibilis (6 times), licit, magistas, manit, mercis, 
oportit (several times), obmuiisco, onisiphorus, pinguido, peripsima, penti- 
custen, paraclitus, profitia, redigerit, reuiuisco, senix, scilieit, xddelicit, uilisco, 
uiri, etc. 

e for i : ancella (4 times, Italian), accepio, agonezari, aures, bestea, crededi, 
collego, dilego, dedici, dessideo, eretis (twice), excesseremux, homecidium, intren- 
secus, porrego, possedeo, praesteti, perdedi, reddedi, sterelis, suscepio, tradedi, 
ueletis etc., and particularly in verb forms of second person plural : lapidastes, 
legistes, recessistes, correxistes, proba.'ites, fecistes, coepistes, seruetes, uelites, 
suffertes, coegistes, habuerites, etc. 

for u : absordus, adolor, adolatio, emolatio, consolo, costodio, copolo, capi- 
toluin, curso, disctpoliis, luxoria, popohis, stimolus, v.olocres (twice), nabochodo- 
nosor, uidto, stodio, oror, saecolo, cocurrem, testimonio, commodo, com, actos, etc. 

u for o : apostidus, apostidatus, auditur, diahulus (nearly always), cogwu.sco, 
consuletur, furtunatus, humicidium, lucutus, hirica (twice), mundu, nundum, 
praeuarieatur, prumptus, pulire, pulluere, penticiisten, parabula, scurpio, sub- 
rius, nullus, sohis. 


y for i: dygamus, aelymosina, hytoria (sic), hyspania, paralypomenon, 
paraclytus^ sapp{-ff-)yra, synapisy syna^ satellytibus, syhilla. 

i for y : azima, hiperbatum, ipocrisis^ idolotitus, misterium, neophitus, pro- 
selitus, praeshiter, sybilla, syntiche. 

ae for e : conpraehindo^ caena, caeler, cotidiae, depraecor^ depraessus, aepis- 
tula, aecclesia (nearly always), aeuu, aerubesco, aelymosina, elymosina, aepvlae, 
aeloqtientia, aebrietas^ aesca, aegredior, aegere, edo, aelisaeus, faemina, haebreus, 
inrepraehinsibilis (6 times), interpraetor, laeuissimus, laquaeus, paenitus 
praessura, praeces^ praetium, praetiosus, p)-aesbiter, quaem, quaerella, reprae- 
hindo, saeueritas, terraestris. 

e for ae : adhereo^ aggeus, egrotus^ enigma, emolatio, ledo, letus, longeuu , 
meror, penite7itia, prestetit, etc. 

t (intervocalic) for c (intervocalic) : amititia, cotispitio, capatior, men- 
datium, ojffitium, pernitiosus, solatium, sotius, speties, initiam. 

c (intervocalic) for t (intervocalic) : diucius, hospiciicm, ospicium, malicia, 
preciosiis, perdicio, pacientia^. 

Much more significance attaches to the following : 

qu for c : aliquius, quum (at least five times), quoaequare (twice), quur (eight 
times), quoheres^, quoram, loquutus, quuius (four times), sequuntur, sequu- 
turus, iiersequutio (twice). 

c for qu : alico, corundam (twice), co, eondam. 

g for c : apogryphus, collogo, gregus, obsegro (seven times), uagare, negare. 

d for i : cedera, confidens, gradus, dubidare, confideatur, gradanter, 
heredicus (twice), hospidem, refudo, tradidi, tradidurus. These intervocalic 
instances have, I think, more significance than the elsewhere frequent final d, 
found here in quod, inquid. 

t for d : impetit, potest {fo?- prodest), canditatus, sordito, nutus, gratus 
(twice), excitant, expetit, etc. 

b for u : abia, coacerbauunt, minorabit, mutabit. 

u for b : coacerbauunt, inuistigauilis, furaueris, suscitauit, multijjlicauit, 
adimpleuites, habeuit, gubernauit, cessauit. 

Such forms as anuntio, aplaudo, apello (twice), acommodo, 
aministratio, asensus, atemto, atendo, afirmo, suportare must 
I think be gi-ouped with the phenomena just mentioned, as well 
as adibiscor (for adipiscor), obrebit, probrius (for proprius), prae- 
cebit (twice). But to no forms should I attach greater significance 
than to : 

guila (seven times) for gula, and menime (ten times). 

1 These may be due to a misreading of ( in a Visigotbic MS : cf. Lowe, Studia 
Palaeographica (Mlincben, 1910) p. 20 n. 1, p. 51 u. 2. 

2 In Rom. viii 17 has quoheredes : Wordsworth and White's only authorities for 
this speUing are CT, both Spanish manuscripts. 


Guila has recently been produced from a Spanish manuscript 
(Madrid, Bibliot. Acad. Hist. 44 (saec. viii — ix) = Table IG** (twice) 
in C. U, Clark's Collectanea Hispanica^), and Mr Robert Weir 
quotes me gyilosus, guilosus from the ' Abolita ' glossary (C.G.L. iv 
81, 24), which is of Spanish origin (W. M. Lindsay in Journal of 
Philology XXXIV (1918) pp. 268 f.-). The forms just given prove 
that our manuscript W9,s copied from a Spanish original. Lest any 
one should doubt the cogency of the reasoning, I must point out 
that Paris 653 is by no means the only North-Italian product 
which was copied from a Spanish original. Some of the oldest and 
best manuscripts of Isidore of Seville's Etymologiae are of North - 
Italian provenance, for example three Bobbio MSS of the eighth 
century (Lindsay's AKL) and a Veronese MS (Lindsay's Phill.) 
coeval with our own^ Beeson's lists of early Italian MSS of 
Isidore's other works tell the same story ^ Verona still houses 
a Visigothic MS which may have been written before A.D. 732 '. 
Direct communication between Spain and Italy was quite frequent 
in the seventh century®, at which date the original of our MS was 
probably brought there. The remarkable character of that original 
is quite consistent with the conjecture that it was an autograph 
compilation of Isidore himself Sedulius Scottus had a commentary 
on the Epistles in his possession, which he calls 'Isidore'.' Though 
the extracts thus labelled come really from Cassiodorus (Pseudo- 
Primasius), we may attribute this ascription to a current tradition 
that Isidore had really compiled a commentary on the Pauline 

^ Transactions of the Connecticttt Academy of Arts and Sciences vol. 24 (Paris, 
1920). My atteution was called to the form guilae in the above-mentioned MS by a 
review of P. Lehmann in Philologische Wochenschrift xli (1921) p. 324. It is in- 
structive to compare Clark's lists (pp. 100 ff.) with the peculiarities noted on the 
preceding page or two. 

2 See also Th. Birt in Eh. Mas. Bd li (1896) pp. 98 f., referred to by Goetz. 

* See the 'praefatio' to Lindsay's edition (1912) in Oxford Classical Texts, and 
Classical Quarterly v (1911) p. 46; also C. H. Beeson, Isidor-Studien (Miinchen, 
1913) pp. 9ff. 

^ Isidor-Studien pp. 23 f., 29, 45, 66, 74, 127 f. 

^ Bibl. Capit. lxxxix (Clark, Collectanea Hispanica pp. 63 f.). 

* Traube, Vorlesungen und Abhandlungen ii (Miinchen, 1910 [dated 1911]) p. 21 ; 
Wilmart, Bull, d'anc. litt. et d'archeol. chret. iv (1914) p. 187. 

'' See below, p. 338. 


The scribes, as we have hinted, were very careful; but they 
had one failing not uncommon among scribes, the omission of 
syllables, especially where two similar or identical syllables come 
together in a word. I have noted these examples : homi(iii)hus, 
C07itend{end)i, prophet{et)is, {de)leo, cred{id)istis, si{di)gnum, st{at)e, 
hort{at)'ur, aedif{ic)ationem, sollic{it)ior, dil{ig)it, oboe{die)ntibus, 
habun{dan)tius, infir{mi)tate, euan{ge)lizo, tolle{re)tur, pecca{ta), 
beat{it)udo, huma{na), de{ne)gemus, restau{ratu)m, mand{at)orum, 
libid{in)em, euang{el)io, nit{it)iir, ex{em)plum, ostend{end)o. 

There are very few examples of the opposite mistake. I have 
noted these: abs\tiii\entibus, adlin]uenticius, eiiipt[at]ionis, ambi[gi]t, 
bapti{s)mu'm\ii7)i\, inmortali[tal{]tatem. 

The kernel of the commentary is of course the original Pelagius, 
but it has become greatly expanded in various ways, in the case of 
all the thirteen epistles except First and Second Timothy, Titus 
and Philemon. In the first place, it contains a number of additional 
short notes the style and language of which suggest that they 
are by Pelagius himself, and that they go back to the author's 
own copy, to which he added notes from time to time. This author's 
copy must somehow have come into the possession of our Spanish 
compiler. He was not, however, satisfied with the increased size 
of the original commentary, .Being in possession of a Pseudo- 
Jerome also, he has added a large number of the new notes he 
found in it, in a text purer than we can otherwise find, because 
quite independent of the corrupt Insular tradition to Avhich we 
owe our copies of Pseudo-Jerome. From the lists in my second 
chapter the reader will get a fair idea of the type of both these 
classes of additional notes ^ A third class of note consists of 
unacknowledged borrowings from Jerome Against Jovinian on 
1 Cor, vii. But the most interesting of all the accretions are a 
number of long extracts quite unsuited in character to the glossarial 
Pelagius as originally composed. These are the following : 

(a) Verbuin caro factum est. Sic accipimus quod — esse non 
desinit [f 5 r, v: anonymous-]. 

1 Pp. 37 ff., 46 ff., 51, 59 f, 

2 Published in Proc. Brit. Acad. vol. ii pp, 435 f. 


(b) De numero apostolorum quaeritur — meritum repperitur 
[f. 6r, v: anonymous^]. 

(c) Hieronimus : Ecce quod non unit — implere non posse. 
Pelagius : O uocem temerariam — uirtutum uincit exemplo [ff. 
41 V — 42 v: an extract from genuine Jerome followed by a long 
extract from the third book of Pelagius's De Libero Arbitrio, of 
which only the last part was hitherto known, and that from a 
quotation in Augustine, De Gratia 39 § 43"^]. 

(d) De eo uero quod solent dicere — anterius posteriusue non 
recipit [ff. 108 v — llOv: an anonymous tractate on the Divinity 
of the Son, undoubtedly by the same author as (/) and (h), this 
author being probably either Fulgentius of Ruspe (ob. A.D. 532)^ 
or some other African of his date and circle]. 

(e) Hieronymus : ' qui nouit,' inquid, — quaecumque uultis, ilia 
facia tis. Pelagius : ut autem ex hoc quoque — ut ultra non seruia- 
mus peccato [flf. 191 r — 192 v: an extract from genuine Jerome, 
followed by a long extract from Pelagius's De Libero Arbitrio, 
otherwise unknown*]. 

(/) Non autem ignoramus quae — uoce absolutissima pronun- 
tiatura [fif. 221 V — 224 r: an anonymous extract, by the same 
author as (d) and {hy]. 

(g) Hieronymus : ' unde et apostolus,' ais, — non esse conten- 
tum. Pelagius : si tuam hoc quoque loco — mihi corona iustitiae 
[flf. 228 r — 229 r: probably from the same work as (c) and (e)^]. 

(h) Quod autem eiusdem sit — credere uelle quod non lego 
[flf. 242 V — 247 v: an anonymous tractate on the Holy Spirit, by 
the same author as (d) and (/)']. 

These and all other additions to the original extent of the 
commentary have already been copied out for press, but their 
publication as part of the present work is inevitably deferred till 
happier times. 

1 Published in Proc. Brit. Acad. vol. ii p. 436. 

2 Published in J.T.S. vol. xii (1910 — 11) pp. 34 f. : the location of the extract is 
due to Dr F. Loofs (private communication of 12 Aug. 1912). 

3 Published in J.T.S. vol. xvii (1915—16) pp. 133—136. 
^ Published in Proc. Brit. Acad. vol. ii pp. 437 f. 

s Unpublished at the time of writing. 

« Published in Proc. Brit. Acad. vol. ii pp. 438 f. 

7 Published in J.T.S. vol. xiv (1912—13) pp. 482—488. 


The numerous doublets in the commentary are best explained 
by the supposition that the compiler used both a pure Pelagius 
and a Pseudo-Jerome, without taking sufficient precaution against 
duplication. Examples of this occurrence are these : 

In Rom. V 15 — 16 Adam solam formain — exemplum uiuendi 
(see pp. 38 — 39). It really belongs to v 16, but Ps.-Hier. has it 
also under v 15: V gives it in both places. 

In 1 Cor. xiii 2 Hinc intellegitur magna fides esse quae in 
euangelio grano synapis coniparatur, quod non propter modicitatem, 
sed propter integritatem uel uigorem appositum est. votandum sane 
quod inter cetera etiam omnis fides quae montes transferal, sine 

caritate non prosit unde intellegitur perfectam fidem grano 

sinapis comparari, et hoc non propter exiguitatem, sed oh integri- 
tatem eius appositum notandum quod inter cetera etiam omnis 

fides sine caritate non prosit. 

In 1 Cor. xiii 8 — 4 Sed non propter deum, sed propter gloriam 
fiat humanam. uel certe siquis in ipso martyrio aduersus fratrem 
retineat iram, contemnens eum qui iussit nos maliciam proximi 
obliuiscere, et in ipsa etiam morte praeuaricatur existens....\^vQt 
clause of verse 4]. Quis non propter deum, sed propter gloriam 
fiat humanam. uel certe siquis in ipso martyrio aduersus fratrem 
retinet iram, contemnens eum qui iussit nos malitiam proximi 
uhliuisci, et in ipsa etiam morte praeuaricatur existens [follows now 
second clause of verse 4]. Similar cases occur at in 1 Cor. xi 2 — 3 ; 
XV 24, 25 ; in 2 Cor. xii 12 ; in Phil, iiii 3 ; in 1 Thess. iii 5. The 
infrequency of these cases shows that the scribes were on the 
whole very careful to avoid the snares into which the abundance 
of their material might lead them. 

A very considerable number of passages have been rewritten : 
I have counted some fifty. The variations will be recorded in my 
critical apparatus, but a few may be selected here to show their 
nature. Sometimes they are mere variations of language ; in 
other cases the differences are more serious. The attentive reader 
will have observed that on Rom. v 14 — 15 this form of the Pela- 
gian text lays stress more than once on the responsibility of Eve', 
while the others leave her alone. 

^ See pp. 47 f. above. 
S. P. 17 


In 1 Cor. iiii 20 
(Original Text.) (Paris 653) 

Hie ostenditur nihil prodesse doc- Ostendit hie nihil prodesse absque 

trinam absque iustitiae operibus ei iustitiae operibus seientiam et doc- 
qui non faciat quae facieuda per- trinam, nouum scilicet testamentum. 

In Gal. ii 10 
Hoc est quod in omni paene scribit Hoc est quod in omni pene epis- 

epistula. tula gentes sollicite commonet et* 


In Eph. iii 8 

' Minimo ' tempore, non labore, [Tempore, non labore in the margin 

simul notandum quod inter sanctos only.^ Est ergo minimus et maximus 
se humiliando fecerit gradus. inter sanctos : quamuis enim se humi- 

liauerit, tamen est res in qua se 
humiliat imus quisque. 

quas sensus humanus adprehen- 'inuestigauiles' humano sensui sine 

dere nisi reuelatione non praeualet. reuelatioue diuina. 

In Eph. V 5 
Contra illos agit qui solam fidem Contra illos agit qui solam fidem 

dicunt posse sufficere. dicunt ducere ad regnum posse caelo- 


In Eph. V 7 

Si in peccatis indicium desperan- Si in peccatis participes fueritis 

tium participes fueritis, eritis et in dei indicium non timentes, eritis et 
poena consortes^. in poena consortes^. 

In Eph. vi 15 

Cum fiducia incedentes intrepide Ut sicut calciati pedes fiducialiter 

praedicate. ambulant, ita et uos intrepide prae- 


In 1 Thess. iiii 10 — 11 

Hoc est, etiam ignotos quosque Etiam peregrinos amando. Sine : 

diligatis. Manibus ojjerando, cum accipiendi 

habeant potestatem. 

1 Commomt et is my conjecture for commoneret of the MS. 

" There is perhaps some primitive corruption here (read dei spernentium for 
denperantium?), but of. iiii 19: Cassiodorus's pupil has neither iudicium desperantium 
nor dei iudicium non timentes. but has all the rest! 


Some at least of these alterations seem to be by Pelagius 
himself. The character of such passages, and the style and 
language of many of the additional notes, as is mentioned above, 
have led me to the conclusion that the basis of the text in this 
wonderful manuscript was the author's personal copy revised by 
him and added to from time to time. 

That this copy is based, at least in part, on a continuous codex 
of the Epistles apart from the Pelagian original text and com- 
mentary, is proved by a number of instances of bad division of the 
scripture text. (Pelagius himself always made his division at 
suitable points.) Examples of this characteristic are these : 

In 1 Cor. xi 23 Quoniam dominus lesus in qua node. lam 
hinc quasi oblitos commemorat quam magnum sit huius mysterii 
sacramentum. Tradebatur accepit panem. lam passurus etc. 

In 1 Cor. xii 15 Si dixerit pes : Quoniam non sum. Pes eccle- 
siae dici potest, qui ad intercedendum egreditur. Manns, non 
sum. Manus ille qui operatur. De corpore, non ideo non est de 
corpore. Dicebant etc. 

In 1 Cor. xvi 18 Refecerunt enim. Circa meum obsequium. 
Et meum spiritum et uestrum. Meum spiritum etc. 

In Gal. i 15 Cum autem placuit ei. Non supra omnes, ne 
iactantiae putaretur. Qui me segregauit etc. 

In Phil, i 19 — 20 Et sumministrationem spiritus lesu Christi 
secund^im. Quia ad meam omnia salutem gubernat. Expecta- 
tionem et spem meam. Quia de omnibus etc. 

In Phil, ii 15 Et simplices sicut filii dei sine repraehinsione 
in medio nationis. Sicut uos deus fecit, considerate enim cuius 
filii sitis, quia cum in omnibus purus ac sanctus sit, filios non 
potest habere degeneres. Prauae et peruersae. Quae omnem etc. 

In 2 Thess. ii 7 — 8 Donee dimidium fiat et tunc reuelabitur 
ille iniquus quern dominus lesus interficiet. Regnum quod etc. 
Spiritu oris sui et destruet. Celeri imperio etc. Inlustratione etc. 

It has been already noticed in chapter ii that a number of the 
notes are given in the margin \ These notes are generally genuine, 
and belong to the commentary in its original form. It does not 
seem possible to conclude that they are taken from a separate 
codex. It appears to me that they were put there simply to save 

1 Pp. 47, 59 f. 



space. They are in nearly every case very short notes that do not 
encumber the margin unduly. It is possible, however, that they 
were overlooked at first, and added afterwards. 

I have noted over two hundred places where the biblical text 
is ' interesting,' or in other words differs from the Yulgate. An 
extended examination of these is not possible here, both from con- 
siderations of space and lack of the necessary materials for com- 
parison. Already in the second British Academy paper^ I had 
occasion to compare the biblical text of our MS in Romans with 
that of other authorities, in over forty passages. The net result 
goes to show that out of forty-three passages where the Reichenau 
MS differs from the Vulgate, in twenty-nine it is supported by the 
Balliol MS, and in twenty-five by the Paris MS 653. 

Here I will take Romans as a whole, and compare its readings 
with the materials collected by Dr H. J. White. If my numeration 
be correct, there are in all 333 differences, great and small, between 
his text of the Vulgate and the text furnished by the Paris MS 
653. In the following cases the Paris MS furnishes a reading 
which appears to be hitherto unknown : 


Paris MS 653 


i 10 


om. semper (so Ball.) 



faciunt ilia 

ea faciunt 

ii 5 

cor sine paenitentia 
in (corr.) geiites 

inpaeuitens cor 
inter gentes 

scriptum est enim 

sicut scriptum est 



in circumcisione 

iii 25 

fidei per fidem 

per lidem 

iiii 4 

contra spem naturae - 
in spem potentiae dei^ 
add sicut stellae et arena 

contra spem 
in spem 


non considerabat 


emortuum oyn. 


V 18 


sicut om. {add mg) 


vi 16 



vii 7 




est alt. om. 


viii 19 

filiorum om. 








1 Proceedings, vol. vii (1915—16) pp. 271 ff. 

■•* naturae is perhaps a gloss which lias got iuto the text. 

^ potentiae dei is perhaps a gloss. 





Paris MS 653 


viiii 11 qui cum 

cum enim 

21 potestatem om. 


32 offendit 


X 4 credeudi 


15 quomodo 

quomodo uero 

19 moses primus 

primus moses 

20 autem om. 


palam apparui 

inuentus sum 

inuentus sum 

palam apparui 

xi 12 mundo 


23 et illi autem 

sed et illi 

25 nolo autem 

nolo enim 

ex parte 07n. 

ex parte 

26 et om. 


34 eius consiliarius 

consiliarius eius 

xii 5 corpus om. 


20 si sitat 

si si tit 

xiii 1 estis 


5 stote 


9 in hoc uerbo om. 

in hoe uerbo 

12 appropriabit 


xiiii 6 et 1° om. 

et X 

XV 6 imo ore om. 

uno ore 



12 exsurgit 


19 jin uirtute signorum et 

jin uirtute signorum et 

Iprodigiorum om. 


21 quia 

sed sicut scriptum est 

24 spero quod praeteriens om. 

spero quod praeteriens 

uobis alt. om. 


xvi 5 epemen 


10 aristobolim 


17 discensiones 


Eliminating these fifty- four passages, we find that there are 
279 differences to which parallels can be produced. I have gone 
through the Wordsworth- White apparatus and counted the authori- 
ties for these, in order to discover with which of them our Paris 
MS is most closely related. It is obviously necessary to confine 
our attention to those that are extant for the whole Epistle S but 
I have included Augustine, because a fairly complete copy of the 
Epistle could be constructed from his ^vri tings. In paying regard 
to the readings of Ambrosiaster and Origen-Rufinus, it must be 
remembered that neither is as yet edited according to modern 
critical requirements, and that the same is true of a large portion 
of St Augustine. The authorities are arranged in the order of 

^ I have iguored e and / because of their well-known connexion with d and g 


their closeness to Paris 653, the number of agreements being 
added in each case. 

d (the Latin side of the Old- Latin Codex Claromontanus, saec. 
VI, which appears to have been contaminated with the Vulgate in 
Romans and the other long Epistles^): 150 agreements, of which 
79 are with cZ, 62 with d*, and 9 with d?. 

L (Paris MS 335, saec. ix-, written in Beneventan script, therefore 
native to S. Italy or Dalmatia): 109 agreements, of which 78 are 
with L, 10 with L*, 3 with L\ 13 with L-, 2 with L^ and 3 with L^. 

D {Book of Armagh): 105 agreements. 

g (the Latin side of the Old-Latin Codex Boernerianus) : 89 
agreements, of which 85 are with g, 3 are g^, and 1 is g°. 

F (codex Fuldensis, saec. vi): 89 agreements, of which 66 are 
with F, 21 with F*, 1 with F^, and 1 with F'^. 

O (codex Oxoniensis, Laud. lat. 108, saec. ix): 79 agreements, 
of which 39 are with O, 36 with 0*, 2 with 0*=, and 2 with O*. 

Ambst. (Ambrosiaster's lemmata): 79 agreements. 

Z (codex Harleianus 1772, saec. viii — ix): 77 agreements, of 
which 29 are with Z, 31 with Z*, and 17 with Z^ 

Orig. (the lemmata in Origen-Rufinus): 71 agreements. 

Aug.: 69 agreements. 

W (codex Sarisburiensis, saec. xiii): 66 agreements. 

The others maybe mentioned more summarily: M (56); H (55); 
T (55); U (52); G (50); C (49); R (49); (46); B (44); K(44); 
c (42); V (41); A (25); dem. (20)^ 

We saw reasons above to conclude that our Paris MS comes 
from a Spanish original. It is therefore somewhat surprising to 
find that the Spanish Vulgate MSS C and T are not represented 
by more readings than they are. The large number of agreements 
with D is gratifying, seeing that we have shown this MS to be 
nearest of all to the text used by Pelagius himself ^ The great 
preponderance of agreements with d shows that there are many 
early European elements represented: this type of text is as old 

* Study of Ambrosiaster Tp. 214. 

2 Lowe, Beneventan Script (Oxford, 1914) p. 356 and passim. 

^ Note also that the Spauish authorities m aud t, which cover only a part of the 
text, have 24 and 20 agreements respectively; gue (8), r (4), r^ (3), p (1). If m and t 
were complete, this would give about 80 with the former, and 44 with the latter. 

■* See chap, iv p. 126 etc. * 


as Lucifer of Cagliari (•f-371). The very large number of agree- 
ments with L is the real surprise of our investigation. I cannot 
find that this MS, written as it was on Italian soil, had anything 
but Italian ancestors. The next MSS on our list, g and F, are 
also Italian in text. It would therefore seem that the text of 
Paris MS 653 represents the Pelagius text, after it has been 
brought into partial harmony with an Italian Vulgate text 
retaining many Old-Latin elements. This text was not closely 
related to the Cassiodorian Amiatinus. It may be, therefore, that 
we are really dealing with a text that, though Italian in origin, 
was actually in use in Spain. Yet the possibility must be kept 
open that our text belongs to the locality to which the MS itself 
belongs, namely the district of Verona. 

The value of this manuscript for the restoration of the original 
comments by Pelagius is very great. This is clearly shown by 
passages already adduced by Zimmer and Hellmann for a different 
purpose, Paris MS 653 having been quite unknown to them: 

Rom. i 3 extinxit ABV Cassiod. Sedul.: exclusit G, excladit 
H (cf in Eph. vi 16). 
21 recedens AV Sedul. : recedentes BGH Cassiod. 
29 intentio ABVH Isid. Sedul.: contentio G: animo- 
sitas Cassiod. 
iiii 1 lit quod in initio constiterit, id habeatur in toto, 
read by AB ; also by V, except that V has con- 
stituerit with Hj : H has adhibeatur for id 
XV 21 uidetur ABV Cassiod. Sedul.: ostenditur GH. 
xvi 24 commemoraret AV Sedul.: commemoret BH.^: com- 
moneret GHj. 
1 Cor. xiiii 1 profetetis A {def. B)V Cassiod. Sedul. : proficiatis GH. 
Eph. iii 18 eligant (A)BV Cassiod. Sedul: diligant GH. 

19 perficiamini ABVGHg Cassiod. (def. Sedul.): efficia- 
mini Hj. 
Col. iii 10 similare AVH* : similari Cassiod. : simulare GHj 
co7'r. Sedul.: simulari B. 

An even better test is to choose places where some of our 
leading MSS are wrong, and see what help we get from V. 


Eph. i 17 nouerat enim earn (i.e. sapientiam) adiutricem omnium 
esse liirtutum AHj Sedul. 
matrem BVG, C {ex corr.). 
matrimonium Hg. 
matrem is clearly the original reading, which was purposely 
modified to soften the hard statement. 
Eph. i 18 si sciretis ad quaiitam spem uocati estis, omnem spem 
saeculi facile contemneretis, et si diuitias here- 
ditatis dei uideretis, omnis terrena uohis horrebit 
hereditas AGHj (exc. quod contempnitis Hj). 
scieritis...contemnetis...uideritis BVH, Cassiod. Se- 
dul: sordebit BV (noii Ho) Cassiod. 
iiii 7 qui ad quam gratiam se aptauerit, ipsam consequitur 
consequetur V. 
consequatur H.2. 
There can be no doubt that V is right here, against all others. 
Eph. iiii 22 qui pristinos errores desiderat huius mundi ABVG*M. 
disserat Rra 1, S. 
dixerat E. 
deserat N, R co7^r. 
et desideria C. 
deserit G (ex corr.). 
repudiat Cassiod. 
We have here to do with a primitive corruption des[id^erat, 
which part of the second family of Pseudo-Jerome MSS has rightly 
emended. It is quite clear that even the Cassiodorus copy was 
corrupt, for his pupil has altogether ignored the ductus litterarum. 
Phil, ii 7 discipulis autem seruiendo A. 

discipidis etiam seruiendo V, rightly, 
seruiendo Hi (seruando R). 
0711. BGHj Cassiod. Sedul. (cum toto contextu). 
2 Tim. i 15 philegus AG phylegus V. 
filetus B philetus N. 
figelus E Cassiod. phygelus R phigelus MC (= D) 

fugulus S. 
phylogelus R corr. 
The common corruption in AGV here seems a compromise 
hetween Jiletus and Jigel us. 

vl] description of mss 265 

The Pseudo-Jerome Manuscripts (H) 
The Pseudo-Jerome manuscripts divide themselves into three 
classes : 

(a) The iininterpolated or almost uninterpolated class, to luhich 
B and the Merton MS belong. 

It is impossible for us to say whether the name of Jerome has 
adhered to this form from the very first or not. In favour of the view 
that it has, is the reference in the well-known chapter of Cassio- 
dorus, who distinctl}^ says that he had left a copy of a brief com- 
mentary which some attributed to Jerome^. This can hardly have 
been anything but a Pseudo- Jerome of some sort-. There is also 
some relationship'^ between B and the longer, what we shall call the 
Irish, recension of Pseudo-Jerome (Hg), suggesting that Jerome's 
name was attached to the B class from the first. There is, however, 
the other possibility that the B and Merton form got the name of 
Jerome added to it because some mediaeval scholar saw its kinship 
with the longer form that he had come to know under the name 
of Jerome. Whichever of the two views be the right one, it is 
convenient to treat the B form as something quite apart, and to 
apply the name Pseudo-Jerome only to the other two classes. 

These other classes have a common root with the St Gall MS 
and also with one another. A signal proof of the close connexion 
of the St Gall MS with both, is a case of a repeated note observed 
by Dr Alfred J. Smith in the published Pseudo-Jerome. The 
following words occur twice in all these MSS, first as a note on 
Eom. xi 17, and second as a note on Rom. xi 24: 

Item: [Hoc] Contra naturam gentilem populum insi- 
tum dicit in radice[m], hoc est, in fide[m] patriarcharum, 
'et non secundum naturam arborum quasi 
insertum proprii generis fructum ferre, sed boni- 
tatem radicis sequi in qua[m] insertus es. [o?' insertum est]^ 
As the note properly belongs only to verse 17, it is clear that its 

^ See p. 16 above. 

- It cannot have been the compilation (still unprinted) which is handed down to 
us in the following MSS : Berne (formerly Micy) 344 (saec. ix) ; Paris B.N. 1764 
(saec. X — xi); Paris uonv. acq. lat. 1460 (saec. x); Dublin, Trin. Coll. 254 (saec. xv) (?). 
This compilation is made from the genuine works of Jerome, and the places are 
indicated. It is probably not earlier than Cassiodorus's time. 

-* Shown in the 'packing' of verses in some of the shorter epistles, etc. 

■* Minor variations of text are reserved for my third volume. 


repetition at verse 24 is due to an error in the archetype of the 
St Gall and Pseudo- Jerome MSS. This error could only have arisen, 
I fancy, from the fact that the note was written in the space 
between two columns of a glossed^ MS of the Pauline Epistles, in 
which verse 17 was in the left hand column and verse 24 opposite 
it, in the right hand column. A copyist, not remembering that he 
had copied the note at verse 17, mistakenly copied it again at 
verse 24, and from the faulty copy thus made all our manuscripts 
of Pseudo-Jerome and the St Gall MS come. There can be no 
doubt also that the faulty copy was in insular script. 

How far back can we trace the Pseudo-Jerome interpolations, 
or any of them ? That they, or most of them, come from a Pelagian, 
cannot be doubted, even though the identity of that Pelagian may 
never be discovered. I have sometimes wondered whether Caelestius 
himself was the interpolator: difference of style precludes Julian of 
Aeclanum. But that some at least of the interpolations are veiy 
old is proved by certain passages in Arnobius Junior, the author of 
Praedestinatus. Von Schubert, in his monograph, Der Sogenannte 
Praedestinatiis^, has produced three (or four) parallels to the inter- 
polations in the printed Pseudo- Jerome. They are Praed. in 14 
(Migne, P.L. liii 653 b) = Ps.-Hier. in Rom. vii 18; ibid. (653 d) 
= Ps.-Hier. in Rom. vii 22; Praed. in 2 (634 a) = Ps.-Hier. in 
Rom. viiii 12. About the cogency of the first and second some 
doubt may be felt, about the third none whatever: 

Ps.-Hier. Praed. 

In Geneseos libra dictum est: Duae In Geneseos libro de Rebecca, dictum 

gentes etc. Ergo prophetia non de his est: Duae gentes etc. Prophetia ergo 

est qui secundum carnem sunt lACoh et non de his est qui secundum caiviem 

Esau, sed [et] de his qui futuri eraut nascebantur, sed de duobus populis. 

ex operibus esse boni et mali ludaeorum et gentium 

Some of the Ps. -Jerome interpolations are then older than the 
date of Praedestinatus, and according to the generally accepted 
view the third book of this work, among other parts, is the work 
of Arnobius Junior^. The date is perhaps not later than 439, and 
the place of writing possibly Rome. 

It is probable that there are strata in the interpolations, that 

1 Glossed, I mean, with Pelagian and Pseudo-Hieronymian notes. 

2 Texte und Untermchungen, N.F. ix^"' Bd (Leipzig, 1903) pp. 35 f. 

^ See Morin, Etudes Te.vtes Deconvertes 1. 1 (Maiedsous and Paris, 1913) pp. 316 f.; 
Schanz, Gesch. d. rdm. Litt. iv" Teil (2) (Miinchen, 1920) pp. 533 ff. 


there is this early element, contemporary or almost contemporary 
with the author himself, and that there are one or two later strata. 
The earlier of the later strata is perhaps to be found in the inter- 
polations preserved by Hi. Certainly some of the interpolations in 
H., must be quite late, such as the extract from Jerome's 28th 
epistle to which allusion has been made\ Again, Pelagius himself, 
according to the A text, says nothing of the identity of the 
'propheta' in Tit. i 12, but the B text (with GH\) has 'Parmenides 
siue Callimachus.' This can hardly be anything but an interpola- 
tion from Jerome on Titus ad loc.^: 'Dicitur autem iste uersiculus 

in Epimenidis Cretensis poetae oraculis repperiri Sunt qui putent 

hunc uersum de Callimacho Cyrenensi poeta sumptum, et aliqua 
ex parte non errant... uerum, ut supra diximus, integer uersus de 
Epimenide poeta ab apostolo sumptus est, et eius Callimachus in 

suo poemate est usus exordio Epimenides siue Callimachus' etc. 

It is not so likel}'^ to be taken from Jerome, epist. 70 § 2, 2, where 
he says: 'Paulus apostolus Epimenidis poetae abusus uersiculo est 
scribens ad Titum... cuius heroici hemistichium postea Callimachus 
usurpauit^' Jerome doubtless got his information, at least partly, 
from Origen, who in turn may be indebted to Clementl The inter- 
polation, with the corruption 'Parmenides' for 'Epimenides' is an 
ugly blot on B's text. Investigation may show that there are other 
interpolations of this sort, justifying to some extent the attribution 
of the commentaries to Jerome®. 

The history of the use of the Pseudo- Jerome as 'Jerome' in the 
middle ages would be an interesting topic, were space available to 
pursue it. I will merely call attention to the fact that Abelard 
(t 1142)^, the Glossa Ordinaria^ and Hervaeus Burgidolensis (f after 
1150)" cite it as 'Jerome.' 

1 P. 239, D. 5. 

2 The MSS of Hj agree in the more deeiDly corrupt text 'parmedissidiae challi- 
macus,' to which the first editor gave the form 'Callimachus scilicet.' H, also 
contains a doublet after 'consecuntur': 'Item Epimenides dixit hunc uersum.' 

3 Ed. Vail.'- t. VII (1) pp. 706—708. ^ Ed. Hilberg (C.S.E.L. liv p. 701,1. 
5 References are Athenag. 30; Clem. Strom, i 59, 2; Orig. c. Cels. iii 43 (cf. 

Geffcken, Zwei griech. Apologeten [Leipz. and Berl. 1907] pp. 227 f.). 

8 Cf. the same feature in V (p. 255). 

^ See the index to Migne, P.L. clxxviii. I think Prof. Lehmann first called my 
attention to this fact. '^ Cf. p. 6. 

^ Migne, P.L. clxxxi. Another MS (not alluded to by Denifle, Luther und 
Luthertum- 1 Bd. 2 Abt. [Mainz, 1905] p. 54) is Rome, ValUcell. e 5 (saec, xii). 


It does not seem profitable to fill space with a discussion about 
the Pseudo- Jerome biblical text. It is very doubtful, in fact, whether 
there was any original Pseudo- Jerome biblical text from which all 
the Pseudo- Jerome codices ultimately derive. We shall see that 
R sometimes goes its own way. We also argue that Ho is based 
upon a separate codex Paulinu^, which may have had nothing to 
do with the biblical text that can realh^ be reconstructed from 
the representatives of H^. I may in fact have to content myself 
with presenting the evidence for the various forms in the critical 
apparatus. It seems probable, however, at this stage, that the person 
who first built the exposition in H., round the codex Paulinus of 
which we have spoken, really believed that it represented Jerome's 
Vulgate text, and it will not be without interest to the Benedictines 
to examine this claim, as the date at which it was done was rela- 
tively early. 

The interpolator occasionally airs a slight knowledge of Greek 
by verbal quotations, a practice alien to Pelagius': the passages 
are these: in Rom. viiii 20; in 1 Cor. xii 31; in 1 Cor. xv 31, 51. 

(h) The shorter or Anglo-Saxon interpolated form of 
Pseudo- Jerome (Hj). 

This form, which I call Anglo-Saxon because its oldest re- 
presentative is in Anglo-Saxon script and it has also other con- 
nexions with England, is that which has become in part known 
through Erasmus's edition of Jerome (vol. ix, 1516), and other sub- 
sequent editions of the works of that Father. It is nearer to the 
original Pelagius in various respects than the longer form can claim 
to be. It has the Epistles in the Pelagian order, Phil , 1 Thess., 
2 Thess., Col., and it furnishes no exposition of the Epistle to the 
Hebrews. The practice of 'subnotation,' also, as defined on page 50, 
is maintained throughout. It is, however, defective in two main 
respects. It lacks the true Pelagian prologue to all the Epistles, 
and in its place furnishes a forged letter to ' Heliodorus.' This fraud 
is so far ingenious that St Jerome really had a correspondent of 
this name, to Avhom the epistles now numbered 14 and 60 are 
addressed. There are no arguments to Romans, First or Second 
Corinthians, any more than there are in MS A of the original form. 
' 2 Cor. vii 11 and 2 Thess. ii 16 are hardly exceptions. 


Second Corinthians ends with the stichoxnetvy Scripta de Macedonia : 
uersus DXCI. Galatians is introduced by the Marcionite argument. 
Ephesians begins with the genuine Pelagian argument, as do 
Philippians (here the Munich MS is somewhat curtailed), and all 
the remaining epistles, that to Philemon being slightly curtailed. 
There are frequent interpolations, introduced by Item or Item alia 
{expositio). In this class we not infrequently find Vulgate chapter 
headings intruding, as at 2 Cor. v 15 — 16. At Rom. v 14 — 15, apart 
from one or two notes introduced by the usual Item, there is a 
short note added on verse 14, while for the long incriminating note 
on verse 15 a short note is substituted^ 

Other peculiarities of less moment are shared by the members 
of this class, as will be shown on every page of the apparatus. Such 
are the following readings: 

in Rom. viiii 21 eos true text: eius H^ 

xi 13 uidt ostendere se uult ostendere 

ad ludaeorum salutem ad ludaeorum salatem 

magnopere festinare viagnoperis festinare 

true text: Hj 

(c) The longer or Irish interpolated form of Pseudo-Jerome (Ha). 

The whole character of the MSS of this longer form gives the 
impression that they go back to an original which was at first 
merely a copy of the Pauline Epistles in Latin, but afterwards had 
the Pelagian and other notes inserted, for the most part, at least, 
between the lines of the Pauline text. This view, and this alone, 
will account for the constant inversion of order by which the com- 
ment precedes the verse or clause which it was written to explain. 
As the basis of this recension is a biblical MS, we find in it the 
prefaces and lists of chapter headings proper to such a MS. In the 
two oldest representatives of the class there is also an almost unique 
set of canons which may have formed part of the archetype. Other 
notable characteristics of the class are the normal order of the 
Epistles, Phil., Col, 1 and 2 Thess., which, as we have seen, is not 
the Pelagian order, and the presence of the Epistle to the Hebrew^s 
with a short exposition. In the account of the St Gall MS it was 

1 See p. 3.5 for the text. 


shown that it contains many of the interpolations present in this 
class, but absent from Hj. 

The exact contents are: 

1. [Canones, confined to M and N]. 

1. [Epistulae ad Rornanos causa haec est argument, confined 
to the other members of the class ^]. 

2. Omnis textus uel numerus prologue. 

3. Primum intellegere nos oportet etc. (a variant opening of the 
genuine Pelagian Prirrtiiin quaeritur quare). 

4. Romani ex Tudaeis etc. 

5. Capitula headings to the Epistle to the Romans. 

6. Romani simt in partes Italiae: hi fidem hahentes, etc. 

7. Then begins the commentary proper, with the heading: 

8. Long argument to First Corinthians, beginning Corinthus 
metropolis ciuitas Achuiae est, published by De Bruyne from biblical 

9. Capitula headings to First Corinthians. 

10. Marcionite prologue to 1 Cor., Corinthii sunt Achaici. 

11. Commentary on First Corinthians; 

and so on with each of the Epistles to the end of Hebrews. I have 
already called attention to the passage added anonymously at the 
end of Philemon from one of the genuine epistles of Jerome^, in 
this family of MSS and in the St Gall MS, A partial stichometry 
is found in these MSS, very imperfectly in the later ones. It is as 

^^11°^^^^^ 1 Cor. DCCCXI. 

2 Cor. Dxci. 
Eph. Dxciii. 
Phil. CCL, 
2 Thess. cviii. 

1 Tim. ccxxx. 

2 Tim. CLXiiii. 
Tit. XLii. 

This stichometry is unfortunately corrupt, and is, so far as I know, 

1 See Wordsworth aud White's Kpistula ad Eoma)ios (Oxon. 1913) for the text 
of these prefatory documents. 

■- Revue Benedictine t. xxiv (1907) pp. 257 ff. » ggg p. 239, u. 5. 


not identical with that of any biblical MS published, though it 
must come from the biblical original I have postulated. 

The situation in Romans v differs in this class of MSS from 
that which we have found in the other class. Here also the in- 
criminating passage is wanting, but there is a further double loss 
(not shared by the St Gall MS), which serves as a means of classi- 
fication of authorities. The first loss is that of the words sicut 
Adam (in Rom. v 14) down to, and including, the words eum 
cupientibus (ibid.): the second is that of the words dindttitur am- 
plius (in Rom. v 20) down to, and including, the words iustitiae 
actionem (in Rom. v 21). The first omission measures rather over 
nineteen lines in Migne, and the second rather over ten. In M, 
the vellum of which is at this point quite normal, 13| lines are 
left blank in the first case, and llf in the second. From this fact 
we gather that the missing passages were no longer legible in the 
exem^plar from which M was copied, and that the scribe left blanks 
of adequate length, in the hope that he might be able to supply 
the gaps from another MS of this work. In this hope he was dis- 
appointed. N wants the same passages, but leaves no gaps, and 
the only sign there that anything is wrong is in a much later 
addition at the foot of the page, in the second case only, in which 
the missing words of scripture are given, but without any comment. 
The other three MSS of this family are also without the missing 
portions, but they have adopted the drastic, and at the same time 
much more interesting course of inserting at both places the corre- 
sponding portions of the Cassiodorus (Pseudo-Primasius) commen- 
tary, not however without marginal notes to the effect that the 
passages thus inserted were wanting in the exemplar. These mar- 
ginal notes are absent from the late and degenerate Cambridge MS. 
The text of this class considerably exceeds that of the other 
class in bulk. Many of the additions are present also in the St Gall 
MS, and can meantime be studied in Zimmer's collation of that MS. 
Yet there are also intentional omissions, for example, of comments 
on the eucharistic section of 1 Cor. xi. Perhaps the theologians 
will be able to locate this tendency. 

It is of considerable interest, in view of palaeographical con- 
siderations which will be stated in their place, to point out that 
this family has got a real connexion with documents of Spanish 


origin. The long prologue to First Corinthians to which allusion 
has just been made, is not known outside this family except in (o) 
certain Biblical MSS: Epinal 45 (saec. ix), closely related to our 
Epinal 6; Paris, B.N. 9380 (Bible of Theodulf— a Spaniard) (saec. 
VIII — ix); Bale B. i 6 (saec. x)^: (&) the unpublished commentary 
of Claudius of Turin (a Spaniard), written about A.D. 820, whose 
prologue, as contained for example in the contemporary MS, Paris, 
B.N. 2392, copied by me, consists of the prologue in question com- 
bined with Ambrosiaster. Thus the oldest external evidence for 
this prologue points to Spain as its place of origin. 

Description of the MSS of the Shorter Form, 
WITH A Discussion of their Relationship 

(3) Pai-is, B.K 9525 (saec. nil ex.) {formerly of 
Echternach abhey) (E) 

This manuscript has 222 folia-, with one column and 26 lines 
to the page. It measures 287 x 193 mm. and was written about the 
end of the eighth century. The arrangement of quaternions is for 
the most part normal. The guard-leaf is not counted, and the first 
quaternion is therefore signed on fol. 9 v. Quaternions 10 and 11 
are signed thus respectively • q • X • and • q • XL This is a sixth 
century fashion, and would seem to have been imitated from the 
archetype. The twenty-second quaternion consists of five leaves 
only, but none of the epistle is lost. The following six quaternions 
are, like the preceding, normal, though the twenty-eighth is signed 
on the seventh leaf, because the eighth is only half a leaf: the 
reverse of fol. 222 is blank. 

The contents of the manuscript are as described above. 

On fol. 3r the former Paris shelf mark 'Suppl. lat. 752 A' is 
twice given, a previous (non- Paris) shelf mark having been removed. 
Below a line which has been erased the same page bears in a four- 
teenth century hand, the words, 'Gontinet leronimu in ues xiiii 
epias pauli^.' 

1 See De Bruyne, Revue Benedictine t. xxiv (1907) pp. 257—262. 
- The numerator has mistakenly given it 223, because fol. 106 was misnumbered 
as 107. 

3 The XIIII is a mistake: the MS never had more than thirteen Epistles. 


Immediately following on the close of the text, that is, after 
the words 'Explicit epistola adphylimonem,' and without any break, 
the same scribe continues: 

exsolutio omnium finita est pauli epistolarum quae 
ameriano' papiaui puerculo^ malo atque laborioso 
scripta est atromento^ uertente anno post obitum 
duorum* idest helis^i^ adsalamonis^ regum 
quis'^ hoc opusculum legat cum caritate 
emendet et excusatvim me habeat qui paruo fru- 
itus sum ingenio :7 :7 : — Amen. 

It is clear that this most interesting 'subscription' does not refer 
to the MS itself, but is copied from the archetype. It occurs also 
(with certain slight variations) in the sister manuscript S, with 
which we are next to deaP. The young scribe of their common 
original calls himself 'Merian of Powys' and dates his production 
as belonging to the year after the two (Welsh) kings Elisha and 
Solomon died . If our knowledge of Welsh history were more precise, 
we could date the archetype exactly, but the most that the autho- 
rities can suggest is that the Elisha is identical with a king of 
Powys who lived between 700 and 750". Merian's MS was probably 
executed, therefore, in the early part of the eighth century. 

After this subscription certain words of the text {simul autem — 
uice sit, i.e. Philem. 22) are repeated, in Caroline minuscule. This 
passage is followed by these words, in capital letters: 


This subscription, unlike the other, was observed by earlier students 
of this manuscript, and shows that the book once belonged to Ado, 
Abbot of Echternach, a.d. 796—817". 

Some comparatively late matter is bound up with the MS proper 
at the beginning and the end. I refer to fragments of a treatise of 
scholastic philosophy, which seem to have been written towards the 
end of the thirteenth century. On the blank portions of these pages 
there are various notes in a German hand of the first half of the 

1 ameridiano S. - pierculo S. ^ atramento [ex atromento) S. 

* ex duonum S. ^ peliri S. ^ ad (con: ac) solomonis S. 

'' qui S. ® The preceding notes record the variations. 

8 Monsieur J. Loth in a note on my original publication of the subscription in 
the Revue Celtique t. xxxii (1911) pp. 152 f. 

^^ Traube and Ehwald, ' Jean Baptiste Maugerard ' (Palaeographische Forschungtn 
Bd. in pp. 336—7). 

s. P. 18 


sixteenth century (fol. 125 verso). Monsieur Leon Dorez, to whom 
I am deeply indebted also for a photograph of one page, has very 
kindly deciphered these fragments for me: 

in (lanT) dem erheren Hans i-oii{T) friburg sol mir xx lib. s. d. 

On the verso of fol. 223 (the last), we read in a hand of the 
fourteenth century (perhaps earlier): 

Filius Hennci regis de nomine quartus \ Rex ptier imbellis sed ad hoc utilis 
armis \ Defuncto patre sub presule coloniense \ Anno eui nomen proprio tutore 
manebat \ Qui presid donee puer hie ad sceptra valeret \ Investituras regni pro- 
visor agebat. \ Comperit hie ergo post quam de famine tanto \ trevei'is ecclesie 
pastore suo viduate. \ Usiis consilio minus ut{'i) patet expirante | discreto qiien- 
dam consanguinitate nepotem. \ Nomine Cunonem puero quoque rege favente | 
Sed qui treburice fer[r]e(?) investivit honore. 

I leave it to students of German histor}^ to determine the exact 
references here. For readers of this book the chief interest lies, 
I think, in the word frihurg in the first extract, and the reference to 
the church of Treves in the second, as will appear presently. 

There need be no doubt that this manuscript remained in the 
Echternach library for seven hundred years after it was written, 
and that it did not travel till early in the sixteenth century. 
Erasmus was then busy with his edition of Jerome. It needed very 
little study to observe that there is a relationship between Erasmus's 
editio pHnceps of Pseudo-Jerome and this MS. Being anxious to 
discover whether there was in existence any evidence as to the 
identity of the 'codex qiiidam obsoletae uetustatis, Gotthicis^ charac- 
teribus exaratus etc.^' which was the only copy of Pseudo-Jerome 
in his hands, I naturally turned to Mr P. S. Allen, Fellow of Merton 
College, Oxford, the accomplished editor of Erasmus's Epistles. He 
very kindly directed me to the Amorbach correspondence, preserved 
in the University Library, Basle, and only in part printed. There 
I found three letters which have a bearing on the question, and 
which I must now quote: 

(G 1 15 p. 9°.) 

Sept. XXI {?xxiv) Anno mdxiii. Bruno Auiorbachius in Basel to his brother 
Bonifacius in Freiburg. 

'■ Commetarios hieronymi in paulum accepi^.' 

' The humanists meant by this word 'Middle-Age, barbarous, and in general 
difficult to read' (Traube, Vorlesungen und Abhandlungen i (Munich, 1909) p. 25). 

- See p. 6 for the following words. 

* Dr Bernoulli, at the instance of my friend Prof. Ed. Riggenbach, helped me 
with the decipherment of this letter. 


This may of course refer to the genuine commentaries of Jerome 
on four epistles, and even to a printed edition^ of these, but the 
probabilities are otherwise. It must be remembered that Bruno 
Amorbach signs the preface to the volume containing Pseudo- 
Jerome, and in it makes reference to the Pseudo-Jerome codex. 
It is probable that Bruno copied out the text of the venerable MS 
for press. 

(G II 29, letter 152.) Gregory Reisch in Freiburg to John (and Basil) 
Amorbach, 10 Oct. 151.3. 


'■ felicitatem in domino consequi perpetuam^ amantissime f rater, quid agatnr 
de S. Iheron scire cupio. misi nuper martyrologiwn ipsius in littera antiqua, ad 
quod conferre poteris martyrologium in treueri excopiatum., et sic remittere illud 
antiquwn. initio et nunc Cometariola eiusdem diui Iheron in omnes epistolas 
paidi, quae esse Iheron testatur glossa ordinaria ad Ro. 1, allegans S. Iheron in 
expositione super eamdem epfam. ponit uerba eadem quae reperiuntur in isto 
codice. etiam facias quaeso libru per filios tuos excribi. item ipsum U7ia cum 
martyrologio remittam ad locum unde accepi sub Chyrographo. In quo loco 
continentur Omelie S. Iheron super cantica^ quamquam initium et finis discordant 
ab his quae impresse habentur. quae uere non sunt Iheron sed potius origenis. 
etiam optime uiue. 

Scriptum ex cartusia friburgensi, altera post dyonisii 1513. 

fr gregorius 


(G II 29, letter 151.) Gregory Reisch in Freiburg to Bruno and Basil 
Amorbach, 12 Aug. 15142. 

'■ Prestantissimis urtium liberalium magistris Brunoniet Basilio Amorbachiis 
fratribus in Christo charissimis. 

S.P.D. Venerabiles ac charissimi magistri, attidit impressor nonmdlos 
quaterniones opera et diligentia vestra castigatos. Et quantum quod angustia 
temporis nunc videre licuit, optime placent omnia, placeret autem summopere ut 
ubicunque Iheronimus ab expositione litterali ad anagogenij) et [one word] 
digreditur., hoc ipsum in margine signaretur, sicuti et concordantiae signantur. 
Id ipsum non dubito tmiuersi desiderarent lector es. miror si ego in exemplar i 
nostro praeter [one word] quoque [one word] ex initio (J) ita signari decreueram 
fi^xi insuper ubi verba textus lxx per minutas litterales [one word] est ad ex- 
positionem quae res et utilis est et opus decorat. Ideo coiisxdo ut in antea similittr 

1 Say that of 1497^1498, published at Venice by Johannes and Gregorius de 

- Kindly copied for me by Dr Carl Roth of the University Library, Basle, at the 
instance of Dr Bernoulli. The script is very difficult, and certain words must be left 
doubtful, even with the aid of Dr Roth's tracings. 



fiat. Ego quidquid ad inventarii collecturam pertinet adiutorio patrum meorum 
conplebo. similiter Excopiata commentana in epistulas Pauli ex antiquo colu- 
mine, cum misse fuerint, ad exemplar [one word lemendandas] ciirabo. prior in 
Yttingen^ optivie notus in sancto Gallo de Job similiter prouidehit. 

Cum iam in nundinis [one or two words] bibliothecam in [one word] per- 
currerem, inueni psalterium littera armenica pulchre scriptum. benevalete. raptim 
ex cartusia friburgensi. Sabbato ante assumptionis anno 1514. 
Fr. Gregorius 
p. Cartusie friburgensisJ 

From the relevant parts of these letters it is clear that Erasmus 
and the Amorbachs received help for the great edition of Jerome 
from Gregory Reisch, as is in fact acknowledged in the prefaces to 
the fifth and sixth volumes of the edition itself^. Reisch was at 
the time Visitor of the Carthusians in the province of the Rhine, 
having been previously, if not also concurrently, Prior of the 
Carthusian house of the Mount of St John Baptist, near Freiburg^ 
It is obvious that such a man, interested in Jerome as he was, would 
be of the greatest help in discovering important manuscripts, and 
obtaining the loan of them. From the letters just printed we see 
that he obtained from some place not definitely named by him, a 
' martyrologium Hieronymi in litteixi antiqua' ^xhich. he sent to the 
Amorbachs, and that he shortly afterwards (10 Oct. 1513) sent 
them, from the same place, a ' commentariola Hieronymi in omnes 
epistolas pauli': for both of these he had signed a receipt. From 
the preface to Erasmus's edition of Pseudo- Jerome we know that 
the MS of the Pauline expositions was also in old and difficult 
characters. Now, there was one library, and perhaps one only at 
that time, which contained very old manuscripts of these two 
spurious works, and that was the library of Echternach. Both 
manuscripts still survive. They are: 

Paris, B.X. 10837 (saec. viii in., Anglo-Saxon large minuscule) 
and Paris, B.N. 9525 (saec. viii ex., Anglo-Saxon minuscule). 

The first contains the 'Martp'ologium,' the second the expositions 
of the Epistles of St Paul. The reader will recall that the word 
'friburg' occurs on a flyleaf of the latter MS, and it may be a part 

' This Yttingen or Ettingeu is said to be in Canton Thurgau, near the Lake of 
Constance (P. S. Allen, Opus Epistolarum Des. Erasmi t. ii [Oxonii, 1910] p. 211). 
2 Allen, loc. cit. 
» Allen, op. cit. p. 27. 


of the very receipt signed b}^ Reisch. It will be proved later by 
textual evidence that the former Echternach MS (now Paris, B.N. 
9525) is the very MS from which, and from which alone, Erasmus 
derived the text of Pseudo- Jerome. After Reisch had duly returned 
it to Echternach Abbey, it does not seem to have been disturbed 
till the Napoleonic perio'l, when, as Traube believes, Maugerard 
stole it, as well as its old travelling companion the ' Martyrologium ' 
and many other Echternach books, and sent them to Paris, where 
they were received on 26th December, 1803^ 

The manuscript is written in rather pointed Anglo-Saxon mi- 
nusculesl That they are Anglo-Saxon, and not Irish, is proved by 
the frequency of the rt symbol (= tin-), which is almost the only 
infallible testl If I am right, at least five scribes took part in 
copying this manuscript: the first and second wrote the first four 
quaternions: the second also from f. 50 r etc.: the third scribe begins 
with the fifth quaternion (f 34 r nolens noceo aliena desidero) and 
also did part of f 50 r: the fourth scribe does from £ 57 v (sine 
miseyicordia), but is soon interrupted by the second, after which 
interval he does all f 58 r (beginning of quaternion VIII), to be 
soon superseded again by the second, who resumes in f 58 v (after 
praeciso), only to be succeeded again by the fourth at the top of 
f. 59 r (atque salutem); this alternation continues for some distance 
down to the end of f 67 v: there appear later traces also of a fifth 
and a sixth scribe, and the total number may have been as many 
as eight. The manuscript has also been annotated by an English 
hand of the latter part of the thirteenth or the early part of the four- 
teenth century. The notes are for the most part marginal, and are 
generally titles of the subject-matter. The book has been disfigured 
by the frequent insertion of paragraph symbols and occasional 

^ I have not access to A. Eeiners, 'Les MSS de raucienne abbaye d'Echternach 
conserves a la Bibl. Nat.' (Publications de la Societe historique de Luxembourg xl 
(1889) pp. 13 — 52), or his article in Studien u. Mittheilungen aus dem Benediktiner- 
uiid Cistercieiiser Orden iv (1) (1884) pp. 429 — 432, or to Publications de la Section 
historique de I'Institut Grand-Ducal, vol. lii (1911) pp. 412 — 478, where perhaps 
further information may be found. Traube's article is iu his Palaeog. Porscli. Bd. iii 
p. 331. 

- Two pages photographed in New Palaeog raphical Society, fasc. viii (1910) 
plate 184. 

2 hindis&y, Revue des Bibliotheques, t. xxii (1912) p. 428; Notae Latinae,pTp. 373 f. 


labels like 'textus' and 'glo(sa),' to distinguish text from comments 
The numbers of capitula are only sporadically indicated in the 
margin or the text, the earliest being xxxiil- of Romans (xi 11). 
The lemmata, which are sometimes in neat uncials, have .. , opposite 
them in the margin on the left. A portion of fol. 117 has been 
burnt out, and various torn leaves have been most carefully repaired 
by sewing. 

The orthography of the manuscript is excellent, and division 
of words is regular. The abbreviations are collected below. They 
agree for the most part with those in the Breton Orleans MS 193 
(221) of Canons^ and are paralleled by those in the oldest Welsh 
MSS\ While our manuscript shows one or two traces of its con- 
tinental surroundings, such as eiiis (second symbol), est (first symbol), 
omnis symbols, the main body of the symbols is Anglo-Saxon and 
may well have been copied direct from the exemplar: a few, like 
apos, may go back earlier still. In view of the Merian subscription, 
the argument for a Welsh (Cornish, Breton) stage in its ancestry 
need not be laboured. 

aliquando aliqn (f. 142 v) Cliristus''' xps etc. 

apostohis apostls (ff. 17 v, 44 v), i ' Christi ' (f. 78 r, doubt- 

aposls (if. 25 V, 45 v, 48 r, £^j\ 

^^y^'fP^^^ '^^■.^■^^'.' 'f ^ cum c nobisc 'nobiscum' (near end 
end or line, 56 rin middle) of line f "^05 v") 

aplr 'apostolorum' (f. 17 r) ^^^^^ 3 (f_ g- j.') ' "^ 

apis (t. 49 r) ^^^^^ ^Ig ^^ ^-^^.^^ 'deltas' (f. 101 v) 

apoli apostoh (f 1 1 3 v) ^■^■^ ^^ (g- ^ 25 v, 1 42 v), die 

ap 'apostolus' (f. 124r, ^^-^j-^^,. dr (f. 94 v ^^r) _ 

doubtful) dicunt dfit (f. 125 v), dint (f. 126 v) 

apos 'apostolo (f. 125 v), ^-^^-^ ^i^ (f ^9^ ^.^ ^^^ ^f ij^^) 

apostohs (corr. in2, dominus d us etc. 

f. 126 v), apostoh ^f. gj-,^. ^(f. ii8r, endofline,f. 212r), 
If ei'(f. 1.37 V) 

apl apostolus corr. rec. episcopus epis 'episcopus' (f. 185 v), 

, ,/^\'}?' ^i- ^^^""'i ^, _ 'episcopis' (or '-os') (f. 

autem hr (f. 27 v first, erased), aut 178 r") 

(ff. 57 r, 70 r), au (fl: 61 v, - , • ■ , . 

1^0 J.N e]Ji 'episcopi etc., eps 

carissimi kiTii' (f. 37 v), (perhaps only . , , . '^Pf 2°Pf ' («,^fj- , if^^ 

three times ^/^^^'«^« epist (f. -2 v),epsl f. 193v), 

epl (f. 213 v), epia 

' See the published photographs. 

■^ Doubtless an error for xxxvi of the ordinary numeration. 

3 For which see Lindsay in Zentralbl.f. Bibliotheksic. xxvii (1912) pp. 264 — 272. 
■* See Lindsay, 'Early Welsh Script' (St Andrews Publications no. 10) (Oxford, 

■5 Christiani written in full (f. 91 r). 




esse ee 

esset eet (f. 213 v) 

est e -h (first f. 36 v, end of line 

f. 65 V, crushed in f. 98 r) 
et 7 (not frequent, sometimes crush- 
ed in where et omitted at first) 
explicit expl, explc (f. 193 v), explct 

(f 213 v) 
expositio expo 
fratres ffs 

ff 'fratres' (f. 70 v), 'frater' 

(f. 91 r) 
fras (f. 142 v) 
hie fe (corresponding to d 'deest,' 

f. 67 r) 
hoc h (f. 141 r) 

id est i- (f. 123 r and often after) 
lesus etc. i&s etc. 
lohannes iofi; (end of line) 
Israhel irl (fl". 43 r, 127 v), isrl (re- 
item it 

mem ms (four times), mm 'meum' qitoniam 
(five times) 

J nobis 
( nobis 


noa^ 'nomina' (f. 185 v) 

non n 
/ noster 

nf 'noster' (f. 66 r first), nfm 
! 'nostrum,' nfa 'nostra' 

J niii 'nostrum' (four times) 

■.itester ufa_' viestra,' ufg 'uestrae,' 
uris ' uestris,' urs ' uestras ' 
' _ (f. 97 v) 

omnis oiTis 'omnes,' 'omnis' (corr. 
m2 omis f. 17 v) 
omis 'omnis,' omi 'omni' 
oma 'omnia' 
om 'omnes' (four times) 
Paulus, pau 'pauli' 
per 4_> 

prae p 

prBm ' presby terum ' (f. 
210 v) 

pbfm 'presby terum' (f. 
219 v), pbri (f. 219 r), 
pbris {ibid.) ' presby te- 
ris,' pbfs {ibid.) 'presby- 
teros,' prsbtri ' presby- 
teri' (f. 3 r) 



pro 43 

propter ^pr (ff. 142 r, 143v) 

(^ptf. 42r) 
psalmus psal 'psalmo' (f. 52 r) 
quae q J^: (f 35 v bis, f. 20rm2), q: 

(fl:: 47 V, 48 r, 48 v) 
quam q (ff. 124 v, 126 v, 127 r, 128 r, 

142 V bis)^ 
{quaiido abbreviated only in ali- 

quare q^-e (f. 143 r) 
que q; (f. 48 r), ij; (f. 18 r, end of line), 
q: (f. 47 v^, etc. corr. to q), 7 
(f. 188r) 
q: (f. 188 v) 

q(f 122 V, 142 v), {j (f. 126 v) 
(what is ^(f. 124 r)? The 
true text is quo), q,(ff. 95 r, 
100 r2) 
qd (J, (f. 37 r, eras, et corr. 
quia, 142 V, 143 r), 9 (f. 
130 r) 

quo (the prevalent form) 
reliqua rlq (f 139 v) 
saecidum sclo 'saeculo' (f. 130 r), scla 
'saecula' (f. 204 v) 
secli 'saeculi' (f. 99 r), 
saecli (f 121 v) 
sanctv.s see 'sancte' etc. 

scs 'sanctos' (f 140 r) 
secundum seed (f 132 r), Jf (f. 143 v) 
sed s; (f 64 v, end of line), s. (f. 140 v 

sicut s (£^25 V, 142.r), sic (f. 126 v) 
spiritus sp5 etc. 

spu ( = spiritus gen. sing., 
thrice at least, e.g. f 68 r 
spum 'spiritum' (f. 172 r) 
sometimes written in full 
(f. 30 v) 
su?it st (once end of 1.), so posst 

'possunt' (f 37 v) 
tamen tiu (f 140 v) 
tempore tempor, temp^ (ff. 15 v, 18 v, 

114 v) 
^lel 1 (f 41 v and later) 

ul (105 r and later) ulit 'uelit' 
(f 147 v) 

1 See Lindsay, Notae Latiiiae, p. 218. 

- Also f. 37 r, where erased and corrected by first hand to quia. The scribe evi- 
dently ought to have used this symbol for quod only. 


Syllable Symbols : 

hunt bt rum f (apostolof 'apostolorum,' f. 

con (fF. 141 r, 155 v, 188 r [with 138 r, end of line; uef 'ue- 

flourish]) rum' flf. 141 r, 220 r ; so meaf, 

en m 'men' personaif. (f. 197 r)) 

er t 'ter,' b 'ber,' u 'uer' _ • runt r, rt 

It b 'bit' c^^cit' a 'dit' il 'pit' ^"' t 'tiir' (f. 9 v, 24 v), t (f. 58r), ct 
It D t)it, c cit, a ait, p pit, gg gg .^^^^ ^^ ^2 

V'K.'' l'''*^^''^'^* ''■'■^^"^'''' 43 r etc. etc.i) expanded by 

for credit, f. 19 r) ^^ 2 very often 

m siiprascript stroke, sometimes be- t (?) ff 144 r) 

yond the letter in the manner ^^^ |^. '■\y^^ j \y. 'bus' 

of very early MSS, e.g. etia" ^ .^us' (grad 'gradus,' f. 138 r, 

(^- ^' V , laudand 'laudandus,' f. 102 v) 

n suprascnpt stroke 

The manuscript behind Merian's transcript may not have been 
in a state of great textual purity, but there is still enough good in 
the descendants of the archetype of this family to show that the 
archetype must have been a good text. The Echternach MS, how- 
ever, swarms with errors, and for these Merian is far from being 
entirely to blame, as is proved by the text of the Salisbury MS, 
alike descended from his. There is a mass of errors special to the 
Echternach MS among our codices. Chapters 1 to 4 and chapter 1 
of Second Corinthians are ver}' corrupt. Perhaps the immediate 
original of E had become very much faded, and was badly worked 
over by a later hand. Of all the errors of E perhaps the most 
interesting in its consequences falls to be mentioned when we come 
to the editio princeps, but two others may be mentioned here: 

In Rom. iiii 17 quamuis in principio uocauerit quae non erant, 
et statim esse coeperunt, tamen hie, quia iam non erant eis tempora 
generandi. Such is the true text, and such is the main text of E, 
but the first hand has added a d(eest), and at the foot of the page 
after h(ic) has written the words defilii di (corr. al. man. de) sperata 
conceptione significat. These words are an interj)olation from the 
Cassiodorus (Pseudo-Primasius) commentary at that point, and 
there is no trace of them anywhere else except in the Wiirzburg 
glossed MS, where however the words are : de insperata conceptione 

In 1 Tim. iii 2 si enini onines unit scire quo modo omnibus 
deheant respondere, quanto magis sacerdos, de cuius ore legem ex- 
qmrent....T\ns, is the true text, but the first hand of E reads si 
1 It may be seen in the New Palaeog. Sac. facsimile, no. 184. 


e7iim omnes uota scire... quanto magis sacerdus uel uxore legem 
exquirent, which the thirteenth century corrector alters to si enim 
omC?) uota scire uolunt uel debentC?) respondere, quanto magis 
sacerdos uel ah uxore legem, exquirent. 

(S*^) The Editio Princeps 

Except the spurious letter to Heliodorus the prefatory matter 
in E is ignored by the editio princeps. This fact does not, I venture 
to think, overthrow the preceding argument that E was the manu- 
script used by Erasmus, and it remains to clench the argument by 
alluding to certain textual phenomena. 

In Rom. i 8 (interpolated passage) ed. pr. reads: ostendit quo 
Romanorum fides omnibus ecclesiis cognita sit. What Erasmus 
intended by quo I do not know, but the later editors expand to 
quomodo. E gives exactly what ed. pr. gives, and the true expansion 
is of course quoniam (M qnrn N quo * [* fuit d] RC qm S quod). 

In Rom. i 19 (interpolated passage) is exactly parallel in every 
way, except that here ed. pr. actually reads quomodo. So again at 
Rom. i 32 (interpolated passage); iii 9 (interpolated passage); 
1 Cor. vi 18 (interpolated passage). The first editor seems gradually 
to have come to know what the symbol quo really meant, for, where 
the cross stroke is actually absent and the true reading is quo, he 
expands to quoniam at Gal. iiii 6. 

But the crowning instance occurs at 2 Tim. ii 24 where the 
true text is secundum quod de eo propheta praedixit. For the harm- 
less de 60 E offers doe. This the editio princeps ingeniously alters 
to Doeg. One of the later editors then asked himself the question 
whether Doeg also was among the prophets, and having ascertained 
that he was not, read Doeg [Isaias], and that is what you find in 
Migne. The reference is, of coui'se, to Isaiah xlii 2\ 

It is only just to point out that, despite these errors and 
multitudes of others, the editio princeps is the best published edition 
of Pseudo- Jerome. In proof of this it will be sufficient to refer to 
one or two passages, without attempting to assess the merits of the 
succeeding editions. Victorius and Martianay are accurate enough, 

1 At 1 Cor. X 4 E has paulus edixit, while ed. pr. has the true rending pulchre 
dixit. This may have been arrived at by emendation, or by the consultation of 
another MS. 


and the earlier Vallarsi edition only less so, but the later Vallarsi 
edition on which Migne depends, has many errors. The later editors 
did something in the way of identifying scripture quotations, and 
they improved the orthography somewhat, but Vallarsi is really 
degenerate. Probably the later Migne is also a degeneration from 
the earlier Migne. Here follow some instructive passages. 

Editio Prixceps Later Migxe 

In Rom. ii 9 (interpolated part) 

Idcirco autem ludaeus geiitili 
antefertur, quoniam praeter natu- 

ralem legem et j\ legem scriptam om. legem et ^loysi 


In Rom. ii 12 
Sine lege literae, in lege naturae om. literae, in lege 

In Rom. ii 26 

ideo circumcisio carnis, indiget om. circumcisio carnis, indiget cor- 

cordis, circumcisio autem cordis non dis 
indiget carnis 

In Rom. xii 16 

qui non dolemus de acie nostra qui non dolemus de acie nostra 

uiros fortissimos corruentes^ uiros fortissimos currentes 

In Rom. XV 24 

nulla magnitudo temporis saciat nulla magnitudo temporis faciat 

caritatem caritatem 

In 1 Cor. vii 26 
l^raesentis uitae sollicitudinem- praesentis uitae solitudinem 

Portions of text and notes at 1 Thess. iii 5, Tit. i 15, Philem. 17, 
which are entirely absent from Migne, are present rightly in the 
ed. pr. It must also be observed that ed. pr., for the sake of brevity, 
often omits the latter part of a lemma, and prints 'etc.': the later 
editors have filled this out, probably from a printed Vulgate ! The 
misleading consequences of such a procedure may well be imagined. 

^ Martianay is still right here; Vallarsi introduced the currentes. 
2 Martianay still right. 


(4) Salishury, Cathedral Lihrarrj, no. 5 {formerhj CLVli 70) 
{saec. XT I in.) (S)^ 

This manuscript has 119 folia numbered, but contains in reality 
121, because there is a blank unnumbered leaf, and fol. 36 his has 
not been counted. It has one column of writing and forty lines 
to the page. It measures now 293 x 195 mm., and was written 
about the beginning of the twelfth century. The top and outer 
margins at least have been clipped I The MS has also suffered from 
damp, and is in an unworthy cloth binding of the nineteenth century. 
No traces of numbers now appear on the quaternions, but they are 
quite normal in character. On fol. 119 (flyleaf) there is no w^riting 
except on the recto side. We find there a paragraph mark, the 
upper part of which has been clipped off, and there are also traces 
of the feet of two letters. Below this occurs AMurKAt(r)is, and 
to the right above, these words: H(enricus) rex Angl(orum) et dux 
Nor(mannorum) et Aquit(anorum) et com(es) And(egavorum). 
Will(elm)o de * * *^ bello ca(m)po^ sa (cut off). The particular 
Henry referred to is doubtless Henry II (1154 — 1189)^ The 
curious subscription common to E and S is in this manuscript 
written twice, but on the second occasion the scribe is a different 
(contemporary) scribe. The contents of the manuscript are pre- 
cisely those of E, except for the omission of In 2 Cor. x 8 {et non 
destructionem) to xi 17 (si hie), without any sign: this part may 
have been so blurred in the original as to be illegible. 

The script of the manuscript is the Caroline type usual in the 
south of England at that date, and the whole book appears to have 
been executed by one scribe, with the exception above mentioned. 
The second scribe appears to have wi'itten also four lines on fol. 93 v. 
The orthography is, like that of E, very good**, and it will presently 
be made clear that the manuscript was copied from an early ex- 
emplar in insular script. 

1 I have to thank Eev. J. F. Shepherd, now of Halifax, for assisting me with 
the collation of the MS, which the Salisbury Cathedral authorities most kindly sent 
to Aberdeen twice for my use. 

- E.g. fol. 83 r, only half of the first x of xxviii remains. 

^ Three (or four) letters lost through a tear in the vellum. 

^ This Beauchamp was Sheriff of Worcester (1155 to 1170). (J.A.R.) 

' I have to thank my colleague Prof. W. L. Davidson, for help in the identi- 
fication. " For instance aput is nearly always found. 


It has certain of the insular abbreviation symbols: s; {=sed) 
on fol. 49 V, end of line, a (= aut) on fol. 51 r, Ir (= autem) on 
fol. 50 v; and tepr (= tempore) on fol. 67 r is not inconsistent with 
the theory of an insular exemplar^ The frequent wrong division 
of words favours an early rather than a late archetype. This manu- 
script suffers more than most of our MSS from horaoeoteleuton. 
The scribe has sometimes failed to complete a word he began near 
the end of a line, and has sometimes also begun the word again at 
the beginning of the next line. Where words in the archetype 
proved illegible or incomprehensible, he Avas apt to leave a vacant 
space of the exact length required. 

That the manuscript from which ours was copied was insular 
of the close of the eighth or the beginning of the ninth century is 
made absolutely clear by the following evidence. Autem is fre- 
quently omitted in our MS, because the symbol hr was not under- 
stood: for a corresponding reason we find con- (o) once at least 
omitted, dicimus (dms) once omitted, eiiis (9) often omitted, enim 
(-H-) often omitted^ est (-^) often omitted', hie (h with vertical 
stroke above the shoulder) once omitted, mihi (m) once omitted, 
uel (I) once omitted. The following corruptions also tell their own 
story. Quo (= quoniam), which is really Anglo-Saxon rather than 
Irish, is found corrupted to quo, quod, quomodo, qua (= quam). The 
following corruptions enable us to fix the date of the exemplar: 
ad dnm for ad nfii (in 1 Cor. i 28) 
non (no) for nostro (no) {in Rom, vi 23) 
na {nam) for na {nostra) {in Rom. viii 26). 
These corruptions were impossible unless the scribe had before his 
eyes an abbreviation system which became extinct about a.d. 815''. 
He was not to blame for not always understanding it. After the 
period mentioned, the nra, nfm system held the field. We find also 
(p) for per (p'), plus (pF) for post (p'). The frequent corruption of 
quia to qu^ (q-) and suchlike is due to pardonable ignorance of the 
value of the insular symbol q. We find saecida (scla; for secunda 

^ See Lindsay, Notae Latinae, p. 308. 

2 We also find autem (Ir) for enim (ft), and eiiim for autem sometimes. On this 
confusion see especially P. Lehmann in Philologus lxxiii (1914 — 1916) pp. 543 — 548. 

3 Omission of the continental e is also possible, of course. 
^ Lindsay, Notae Latinae, pp. 148 f. 


(scda) (in 1 Tliess. iiii 11): sclm is an Anglo-Saxon rather than an 
Irish symboP, and if our scribe was accustomed to it in the exemplar, 
the mistake is natural. The sed symbol (s) was also misread by him 
as si, se. 

The confusions between letters point the same way as the errors 
due to misunderstood abbreviation: 

a was misread as u (very frequently), o (about six times), cu, eo, ec, co. 
This was possible only if a had the oi^en form, and the open form belongs to 
early rather than to late minuscule. 

c misread as t (possible already at semi-uncial stage). 

€C ,, „ U. 

m „ ,, nt (may go back to semi-uncial stage), r (once). 

n „ „ h (several times), r (several times). 

nt ,, „ m (may go back to semi-uncial stage). 

r „ „ 71, s (often ; oftener than the opposite), u. 

s „ „ 7- (often), n (once). 

M „ „ 71. 

2 » „ ff- 
The final t is sometimes omitted in verbs, e.g. opta for optat, falla 
{or fallat, and we often find single letters for double, e.g. asero, acuso. 

The immediate ancestor of the Salisbury MS was an Anglo- 
Saxon manuscript of the end of the eighth or the beginning of the 
ninth century. Though contemporaneous with the Echternach MS, 
this ancestor was not the Echternach MS, nor even strictly speaking 
a sister MS. A curious analogy to the relationship between the 
Echternach and Salisbury MSS is to be found in the relationship 
between the other Echternach MS already alluded to (the Martyro- 
logy), and the Ricemarch MS recently edited by Dr Lawlor^. The 
Ricemarch MS, now Trinity College, Dublin, MS A. 4. 20, was 
written in Cardiganshire, South Wales, about A.D. 1079. 

S is, in fact, on the whole, a better MS than E. Taking a few 
places at random, we find S right where E is wrong: 

hi Rom. V 6 an tarn benefico et sancto aliquid pfaeponendum sit, 
where E has aut before aliquid. 

In 2 Cor. i 6 obtinent : obtent E. 
„ „ „ 13 me : meum E. 

1 Though of course common elsewhere. 

2 The Psalter and Martyrology of Ricemarch, edited by H. J. Lawlor, vol. i 
(H.B.S. vol. 47) (London, 1914) especially pp. xiv, xviiiff., xxvff.: 'E is closely 
related to E' (p. xxv). 


S is really independent of the special errors of E. There is also 

some extent of difference where the advantage is on the side of E. 

Neither E nor S can be a direct copy of Merian's manuscript. For 

on the one hand we have found E swarming with errors which are 

not to be found in S, and on the other hand we have found that the 

immediate ancestor of S belongs to as late a date as saec. viii — ix^ 

We shall therefore not be far wrong in constructing the following 


Merian's MS in Welsh script 
(saec. VIII in. ?) 


Saec. VIII med. O 
Insular I 

Anglo-Saxon script ^ E Q Anglo-Saxon script 

saec. VIII ex. 

saec. VIII — IX 

_ Caroline 
(saec. XII in.) 

If we seek to penetrate behind Merian's production to the MS 
which he was copying, we may conjecture that it was a half-uncial 
of some sort of either the sixth or seventh century. 

(5) Munich, Staatsbibliothek, kit. 13038 {saec. ix in.) 
{formerly of St Emmeram in Ratishon) (R) 

This manuscript has probably 390 folia, not counting a fifteenth 
century guardleaf at the end. It has two columns to the page, and 
26 lines to the column. It measures 315 x 235 mm.- Each column 
is rather over 24 cm. long and rather over 8 cm. broad. In the 
opinion of the great palaeographer, Professor Paul Lehmann of the 

^ It seems impossible to say where this ancestor came from, whether from an 
English or a continental centre. I am greatly indebted to Canon Christopher 
Wordsworth of Salisbury for a number of detailed notes as to the connexion of Old 
Sarum with the Province of Rouen and the diocese of Bayeux, as well as with 
Lisieux, Coutances, Bee, Fonteuelle, Caen, Jumieges, Mont St Michel, and St Wand- 
rille. I lack the necessary knowledge to follow up these interesting facts. 

^ The MS was once taller : the binder of saec. xv — xvi has clipped off a whole 
line at the top of fol. 183 recto and verso, and also parts of lines at the top of f. 201 v, 
f. 217 v. 


University of Munich, it can hardly be of earlier date than a.d, 810. 
The Avriting is a large dark Caroline minuscule of the South German 
type. The arrangement of quaternions is far from regular through- 
out. The first and second quaternions are an afterthought, due to 
another scribe, Avho, as we shall see, did his best to remake this 
manuscript into a member of the other Pseudo-Jerome family. 
They bear no signatures, and fol. 14 v is quite blank. 

unnumb. 12 3 14 5 6 7 

I I I I ! I I I 

9 10 I 11 12 13 14 

The codex began originally with what is now the third quaternion: 
X 15 16 17 I 18 19 20 21 

This is signed on the lower right corner of fol. 21 v: I. The next 
is signed II in the same position (fol. 29 v): III (signed in the same 
way on fol. 35 v) is a ternion of very thick vellum. IIII (signed so 
on fol. 45 v) is thus constituted : 

36 37 38 X 39 40 i 41 X 42 43 44 45 

V (signed so on fol. 53 v) is thus made up: 

46 47 X 48 49 j 50 51 52 x 53 

VI (signed so on fol. 61 v): 

54 55 X 56 57 I 58 59 60 x 61 

I I ! I I I I I i I 

VII is normal, but is signed on fol. 67 v, because leaves after 65 
and QQ have been overlooked in the numbering : Villi (fol. 83 v), 
XI (fol 99 v), XII (fol. 107 v), XIII (fol. 115 v), quaternion XV 




(unsigned), XVI (last I partly erased, because XV was unsigned, 
f. 139 v)', XVII (f. 147 v) are all normal: but VIII (signed f. 75 v) 
is made up thus: 

68 69 X 70 71 I 72 73 74 x 75 

X (signed 91 v) thus: 

84 85 86 X 87 I 88 89 x 90 91 

XIIII (signed 123 v) thus: 

116 117 X 118 119 1 120 121 122 x 123 

XVIII (signed 157 v) thus: 

148 149 x X 150 151 I 152 153 154 155 156 157 

XVIIII (signed 165 v) thus: 

158 159 160 X 161 i 162 163 x 164 165 

XX (signed 173 v) thus: 

166 167 X 168 169 1 170 x 171 172 173 

I I ; I ! ^ : 1 I 

XXI (signed 181 v) thus: 

174 175 X 176 177 I 178 x 179 180 181 

I I I 

XXII (signed 189 v): 

182 183 X 184 185 I 186 x 187 188 189 

^ XVI (at fol. 132 a) begins like a fresh codex, and is by a different scribe from 
the latter part of XV. At the very top of f. 132 is 'IIII • pars hruommano.' 


23 is unsigned (197 v): 

190 191 X X 192 I 193 194 195 196 197 

XXIIII (f. 203 v) is a perfect ternion. 25 is unsigned (210 v) 

204 205 206 207 1 208 209 210 x 

After fol. 210 several leaves have been cut out; they bore writing 
by the first hand and marginal additions by the leading corrector. 
They were then cancelled, the corrector writing all. XXVI (signed 
220 v): 

XXVII (f. 228 v) and XX Villi (f. 243 v) are perfect, but XXVIII 
(signed 235 v) is thus made up: 

229 230 231 x I 232 233 234 235 

ff. 244 — 247 are a perfect binion, and f. 248 would seem to end the 
codex proper (end of exposition of Philemon : Hebr. begins on f. 249 r). 
f. 265 v is signed XXX, which indicates that all is confusion at this 
point, the MS being made up of patchwork, the outward sign how 
difficult was the task of altering a MS of the short recension into 
a MS of the longer. The rest of the MS, with which we are not 
really much concerned, is for the most part made up on a regular 
plan, the signatures being found on the following leaves: 273 v 
(XXXI), 281 V (XXXII), 289 v (XXXIII), 297 v (XXXIIII), 305 v 
(XXXV), 313 V (XXXVI), 319 V (XXXVII), 327 v (XXXVIIT), 
335 V (XXXVIIII), 343 v (XL), 351 v (XLI), 359 v (XLII), 

^ 216 is only half a folium, the recto half bearing no writing. 
S.P. 19 


367 V (XLIII), 375 v (XLIIII), 383 v (XLV): 387 v finishes the 

In A.D. 1462 a scribe went over the whole MS, underlining 
black lemmata with red, adding modern chapter numbers and 
headlines etc. He himself furnishes the date of his activity in 
red letters on fol. 387. 

1. Pseudo-Hieronymi in epistulas Pauli. 

f 1 r. In nomine {later addition) INCIP prologvs • IN EPISTOLIS 

PAVLI {eras.) APOSTOLL Omnis textus uel numerus...emendatus 
melior factus. explicit, incipit prologv.s sci hieronimi presbi • 
{red). Primum qugritur quare...(f. 3ra) manentem substantiam. 
explicit, incipit argvmtvm solivs epistol^: ad ROiiANOS. Romani 
ex iudgis...concordiam cohortatur (f 4a) finit. incipivnt capi- 
tvla I. De natiuitate dni secundum carnem...(Li)...uero ipsius 
reuelatio. explicit (f. 5 v a) incipit argvmentvm Romani sunt in 
partibus italiac.ab athenis. explicit argvmentvm (f 5 vb) IN- 
achaig similiter... abepheso. INCIPIVNT capitvla. i. De plenitudine 
diuitiarum...(LXXii) eos qui credunt dno ihu. Then capitula to 
2 Cor., argument to 2 Cor.; argument to Gal., capitula to Gal.: so 
with Eph., Phil, 1 Thess., 2 Thess., Col., 1 Tim., 2 Tim., Tit., 
Philem....luca adiutoribus pauli. EXPLICIVNT, 

f 15 r (beginning the original codex) INCIPIT PROLOGVS IN 

tu4s {corr. al. man.). , .pdicant EXPt PROLOG; IncIp corpvs • EPis: ad 
ROMAN: The MS proceeds like the others of its family, and would 
have ended (f. 248 v a) like them, save for the Merian subscription, 
had not the patchwork process already referred to been carried out. . . 
(f. 249 a) INCIPIT PROLOGVS AD HEBREOS. Haec nos de intimo he- 
breorum...(f. 262a) caesg uictimae non remittunt. explicit ex- 


2. Hieronymus in epistulas Pauli ad Ephesios, ad Titum'. 

Title on the cover: 42. | 57. | in epistolas D. Pauli Coment. | -PS 

' The text of these genuine commentaries appears to be good, if a cursory 
inspection may be trusted. 




{monogram) ■ paper label. Rat. civ 38. Paper label Cod. lat. 13038. 
xxxxii {old). Old wooden boards: clasps goneK 
The following abbreviations occur: 

apostolus apis, apstls (twice), apostls 
'apostolus,' apl 'aposto- 
li,' apostli 'apostoli,' aps 
(corrector frequently) 

autem aut (au corrector once) 

Christas xps 

cum c (uotc, uobc, nobc) 

Dauid dd 

deus ds 

dicens dies 

dicit die, dit (end of line) 

dicitur dr, dif 

dominus diis 

ecclesia ecla (corrector) 

eiios ei' 

eniiii ■|-|' (once only? above the line) 

episcopus epos ' episcopos ' (corrector) 

epistula epis, epla, epl 

esse ee (also ees, eent 'esses,' 'essent,' 

est -h (twice, once above line, once 
by corrector) 

et 7 (once only, where et missed out) 

euangelium euanglo, euglu 'euangelio' 

frater frt (twice) 

gloria gla 

gratia gfa 

hoc h (corrector) 

homo ho 

id est -i- (once, corrector) 

lesus ifis 

lohannes iohaes 

Israhel isrl, irl, isrhl (end of line, 

item it 

meus ms 

mihi m 

nisi n 



ni, nni, no, nis etc. 

nr!s, nra (corrector) 

ntr, ntrni, ntra etc. etc. 

(most frequent)"^ 
utr, utfm, utfo etc. (t twice 

uls 'uestris' 

(nunc n 

[twic t, t (once each) 
ohlatio oblo (end of line) 
omnipotens omps, omnipts 
om7iis onins 'omnes' 

onia, omes, omibus, omis 

omm 'omnem' 

onem 'omnem,' ones 'om- 
populus ppli 'populi' 

post p 

presbyter prbos ' presby teros ' (cor- 
propheta ppfi, ^pS 
propter pp, (ppt 


(qua q 
{quo q 


quae qg 

que q.-q; (expanded once by correc- 
tor) . 
qid g, (corrector alters to qui or q) g^; 
quia q< 
quod qd 
quoniam qrn 

(rarely qniii, quo) 
reliqua relq (rl corrector) 
saeculum scla 'saecula' 

secla, sgcli 
sanctus scs 

scilicet s- (? a later hand) 
secundum seed, scdm, secund (often), 

scd 2 
sequitur seqf, seqt 
spiritus sps 

SM?li S, St 

tamen tm (corrector, should be 

tempore tempf, tempfe 
^ie;^ 1 (rare), ul 

1 The MS was kindly sent to Karlsruhe for ray use in July 1906 : the collation 
was made at Munich seven years later in the Arbeitssaal of the UniversitatsbibUothek, 
by kind permission of the authorities of both hbraries. 

' Traube (Nomina Sacra, p. 225) had noted this speciality of this manuscript. 



Syllable Symbols : 

bis B (uerB 'uerbis,' expanded by ra, ri etc. siiprascript a, i, etc. (sup 

hrcnt bt '"'''"''''^ '^^P'"^'' P^^^ 'P""^^^') 

con (corrector) ^^^^ y 

dum d(mod 'modum,'cremand'cre- ^^^ :^ 

mandum,' etc. see secundum) . 2 

en m 'roeii' ^w ^ ^ 

it osteud 'ostendit' 
mils, mis m', vcq, m^, n' (see eius), 

n; (corrector corrects to 

n'), Tif, bush; pus }p', 

The long i is found usually at the beginning of words, whether the 
letter be vowel or semi-vowel. The ligatured ti is also indiscrimi- 
nately employed. 

These abbreviations are for the most part the regular abbrevia- 
tions of a Ratisbon scriptorium, which was of course under Anglo- 
Saxon influence. It is not very easy to determine exactly which of 
the abbreviations were taken over unaltered from the exemplar. 
That the exemplar was itself insular is hardly to be doubted. But 
a few of the abbreviations, such as apstts, apostls, gla, gfa, suggest 
that at some stage of the transmission it was represented by a 
Spanish copy. 

That there was a Spanish stage in transmission is also suggested 
by the habitual use of-quu- in words like loquutus, consequutus, quur, 
for the usual locutus, consecutus, cur. But this Spanish stage must 
have preceded the insular. The omission of ergo (g) at least twice, 
and of autem (hr) a good number of times, as well as the writing 
ofenim (tt) for autem (hr), are proof sufficient of an insular exemplar 
in front of our scribe. This view is backed up by the confusion 
between r and n which occasionally appears: the confusion between 
a and u is witness to an 'open' a in the exemplar, as we should 
expect at the date when it was written. 

R has many corruptions absent from E and S, for example: 
In Rom. i 24 iudices R; uindices ES: in Rom. i 25 est deus bene- 
dictus R; est benedictus ES: in Rom. ii 4 cum R; cur ES: in 
Rom. ii 11 sibi quid R; sibi quia E; sibi qui S; quidem sibi Hg; 
true text 'sibi' (alone): in Rom. ii 17 uerum esse R: esse uerum ES. 

This MS sometimes agrees with E against S, more often perhaps 
with S against E, as we should expect. Two examples of the latter 
occurrence will suffice: In 2 Cor. v 16 neminem nouimus carnaliter 




circiimcisum — exemplum om. RSH2, hahet E: in Gal. iiii 24 nos 
uero praeter auctoritatem RS; nos uero praeter caritatem E. 

In some cases it appears as if an ancestor of R had rewritten 
the text, for example: In 1 Cor. vi 11 in tali delicto R; ut adsolet 
E; ad haec S: in 2 Tim. ii 24 eiulauit R; clamauit ES. 

Yet R is an undoubted member of the Hj family. In view, 
however, of the many differences from E and S,and the entire absence 
of the Merian subscription from R, it is on the whole probable that 
R is not descended from the MS executed by Merian, but belongs 
to a collateral branch; thus: 




Insular O 


Insular O 

-O Merian (Welsh) 

O Anglo -Saxon 

R (see below) 

The work of the special corrector of R is more properly referred 
to in the part of this chapter concerning the other recension. 

In Biblical text and prologues this MS has distinct points of 
contact with the Biblical MS M (= CIm 6229 saec. viii) used by 
Wordsworth and White in constituting the text of Romans. 

(6) Munich, Universitdtsbibliothek, Cod. MS 
in folio 12 (saec. XV ex.) (r) 

This manuscript is a paper MS of the years 1490 and 1491 con- 
taining 253 folia. It comprises all Pseudo-Jerome as well as the 
commentary on Hebrews, and the genuine Jerome on Galatians, 
Ephesians, Titus and Philemon. Preceding the Pseudo- Jerome part 
is a sort of alphabetical index of topics in the Epistles of St Paul, 
beginning Abrahae filii sunt gat 3 g: then the Oinnis textus, the 
Romani sunt in partibus italiae, the De natiuiiate domini capitula, 


also the prologues and capitiila to the other Epistles as in Clm 13088 

(order 1 Thess. 2 Thess. Col). The Litteris tuis is followed by the 

commentaries on the Epistles (order 1 Thess. 2 Thess. Col.). Much 

space is saved by giving frequently only the initial letters of words 

in verses. I convinced myself that for Pseudo-Jerome at least this 

MS was copied from the older Munich MS. It must of course have 

got the genuine commentaries on Galatians and Philemon from 

some other source. The cover has the following inscription inside : 

'Hunc librum legauit fratribus minoribus dominus Achacius Hais- 

wasser predicator In Elpogen' anno dm 1516.' Another inscription 

reads: '...tunc rector scolarium ratisbone recessurus tamen ad ange 

reminiscere ad susceptum statum ad S Judocum landtzhuete-.' 

I have naturally made no use of this MS in constituting the 



Description of the MSS of the Longer Form, with 
A Discussion of their Relationship 

(7) Paris, B.N. 1853 (saec. vjii—ix') (M) 

Older shelfmarks are preserved as follows ' Cod. Colb. 2065 Regius 

„ , 1853 Jac. Aug. Thuani.' It is absolutely certain that the 

MS belonged in the sixteenth century to the first library of the 
Jesuits at Paris, formed before 1594, which was pillaged by De 
Thou about the end of the sixteenth century^ 

The MS consists now of 255 ff., but once had at least 256. The 
canons at the beginning are arranged in two columns to the page, 
but elsewhere there is one column only to the page. Each page 
contains 25 lines, and measures 295 mm. by 191 mm., while the 

1 In Bohemia, as I learn from Prof. Paul Lehmann, to whom I am also indebted 
for the knowledge that the Munich University Library contains MSS and for the 
decipherment of certain difficult parts of these notes of ownership. 

2 The modern Landshut (62 km from Regensburg [Ratisbon]). 

3 As to the date, various palaeographers have favoured me with their opinions : 
C. H. Turner, last quarter of saec. viii ; W. M. Lindsay, saec. viii; E. A. Lowe, 
saec. VIII — IX ; A. Holder and H. M. Bannister, saec. ix incip. 

■* This information I owe to a kind examination of the MS which Monsieur H. 
Omont undertook ou my behalf. The pressmark of the Jesuit library has been erased. 
Cf. also Delisle, Cabinet des Manuscrits t. i p. 437 ; Traube, Vorlesuncien und Abhaud- 
Inngen Bd. i p. 17 n. 1. 




written part measures 233 mm. by 135 mm. (approximately). Not 
fewer than twenty scribes took part in the production of this 
manuscript, which is for this and other reasons of gi-eat palaeo- 
graphical interest. It is difficult to give the precise arrangement 
of the quaternions because of the manner in which the manuscript 
has been bounds The earliest signed quaternion of which the 
signature is now visible is XIIII on the right inner lower corner 
of fol. Ill V. The existing series is as follows^: 

XIIII on fol. Ill V (right inner 
„ 127 V (middle foot 
„ 135 V ( „ 








ower corner) 

„ 143 v( „ 

„ 151 v( „ 
„ 159 v( „ 

„ 167 v( „ 
„ 174 v( „ 

(leaf cut out between 172 and 173) 
XXIII „ „ 182 v (middle foot) 
aXXIIIJ „ „ 190 v( „ 

XXV „ „ 198 v ( „ „ ) 

Leaves have also been cut out as follows: two between fF. 89 and 
90, one between ff. 244 and 245, one between ff. 245 and 246 (con- 
taining a portion of the Philemon commentary and part of the 
Hebrews argument and capitula), and one between ff. 248 and 249. 
Only in the one specified case does this mean any loss of text: the 
others are merely cases of cancelled leaves. 

The contents are as described on an earlier page, but the title 
of the commentary proper may be repeated here (fol. 12 r) 


Such a grandiloquent title is not confined to the MSS of 
the second family of Pseudo-Jerome. I am unable to produce 

1 It belongs (like Paris B.N. lat. 2709) to a group of MSS bound superbly in 
crimson morocco about the middle of the nineteenth century, but the sheets have 
been so tightly bound that the volume does not open easily and one cannot see the 
inner edges of the sheets. 

- A number appears to have been erased on fol. 7 v. 


another instance of this precise phraseology, but parallels are not 

To the best of my belief, no photograph of any part of this 
extraordinary manuscript has been published. A study of the MS 
from the palaeographical point of view may be commended as a 
most interesting piece of work, wdiich ought to be accompanied by 
a carefully selected series of photographs designed to show the 
writing of each scribe. The lamented Abbe Liebaert had taken 
photographs of folia 12, 23 v (1, 2), 78 v, 94 v, 115 v, 185 v, 205, 
206 V, 222 V, as I learn from Professor Lindsay's list-. The present 
writer possesses beautiful rotographs of ff. 137 r, 159 r, 159 v, 177 r, 
177 V, 178 r, 184 r, 245 v, selected for their textual rather than 
their palaeographical importance. The scribes, with the exception 
of the last, use a pre-Caroline minuscule, which for the most part 
slopes in an extraordinary way, the upper part of the letters falling 
some degrees to the left of the perpendicular. Dr H. M. Bannister, 
to whom I showed the MS, was struck by the crowd of ignorant 
scribes, all taught to write in the same way. He thought that the 
last scribe, who wrote in Caroline script, came perhaps from else- 
where, and filled in the letters with red ink. The question of the 
scriptorium in which this manuscript was produced will be discussed 
after the numerous abbreviations have been recorded. 

apostolus apostl 'apostolus,' apostli, apsl 'apo.stolus,' 'apostoli' 

ap'tti 'apostoli,' apostl, apos 'apostolus,' 'apostoli,' 

ap'tlo 'apostolo,' 'apo- 'apostolo' 

stoli' (f. 226 r), ap'tulu apis 'apostolus,' aplo 'apo- 

apostls, ap'tts 'apostolus' stolo,' aplos 'apostolos,' 

ap ' apostoli ' aplis ' apostolis ' 

aps apost 'apostolus' 

apstolo 'apostolo' aposl, aposli, ap'li, 'apo- 

apstlo 'apostolo,' apstli stoli,' ap'lo 'apostolo' 

'apostoli' apud ap, apd 

apl 'apostolo,' 'apostolus' antein au,,ailt (rare), h- 

' For example, Metz 134, Oxford, Bodl. Auct. T. ii 24 (saec. x — xi) {Corp. Glosii. 
Lat. IV p. xxxv), Paris B.N. 12289 (Fleury), Einsiedeln 131 (saec. x) (E. v. Dobschiitz, 
Das Decretum Gelasianum, p. 16), Chartres 31 (saec. ix) (ibid.), Laon 273 (saec. ix 
ex.) (C.S.E.L. xxm p. 231), St Gall 1.58 (saec. ix) (C.S.E.L. Lip. xv), Brit. Mus. 
Harl. 1772 (saec. viii — ix) (ed. Buchanan), Paris B.N. 12,125 (saec. ix) (Corbie) etc. 
(Origenes' Werke, Bd. v [Leipzig, 1913] p. lix), show such titles as In Christi nomine. 
In nomine domini, In nomine domini nostri lesu Christi. 

2 Copies of these may be obtained from Sig. Pompeo Sansaini, Via Antonio 
Soialoja 3, Eome. 




capitulum etc. cap, capil, capl 
caput capd (on analogy of apud) 
carissimi kmi, kms 'carissimus' 
cetera c&, c&ef, caet, cet 
christianus xpiano 'christiano' 
Christus xps, xpo (sic, f. 109 r) 
Colosenses colosens, collos, colos 
consolationihus consl 
Corinthios corint, chorint, corin, cor, 

cum c 
Dauid dd 
de d (prep, and syll.) 
deus etc. ds etc. 
diaconii {-atus) diacon 
dicere dfe 
dicimur dmr 
dicit die, dit, dt 
dicitur dr 
dicunt dnt 
dicuntur dnr 
dixit dix, dx 
domimis etc. dns etc. 
ecclesia aecla, aeclm 'ecclesiam'; 

aeclae, §ccl§ 'ecclesiae' 
eius ei (once ei, f. 178 v) ei; e; (fF. 
140 V, 198 r) el (often) ej ej; 
ej: ej, (these four in ligature) 

enim "j-J" 

Ephesios efF&e, efphes, effew, ephes, 

episcopus eps, epis 'episcopis,' 'epi- 
scopi,' epm 'episcopum,' 
epi 'episcopi,' epos 'epi- 
episcopm 'episcopum' 
epistida epis 'epistula,' 'epistulae' 
ep 'epistula,' 'epistulae' 
epistl 'epistulae,' 'epistu- 

eple 'epistulae' 

eps 'epistularum' 
epi 'epistulae' 
esse 11, ee, ee 

essem etc. fgm ; g^, fft, eet, eet, 'esset ' 

est e -=- (see hoc) 

et & 

explicit exi^l, explic, expli, exp(?), 

expositio exp, expos 
fratres frs, ff, fras, fPs, M 'frater,' 


frte 'fratre,' fts 'fratribus,' 
fft 'frater' 
Oalatas gala, g5,l 
haec h. (iiec, f. 231 v) 
Hebraeos hebf 
hera (= section^ Lir, hf 
Hieronymi hier, ier, hieron 
hoc h-(f. 211v) 

hoc est h -^ (5 times, Caroline, 
lestis etc. ihs etc., ihu ' lesum' (f. 81 v), 
ihm 'lesu {gen.)' {K 155v, 
190 v)_ _ 
incipit incip, incipt, incipi 
lohannes ioti 
Israhel isrl, irl, isrlfi, isr&l, israfel 

(f 239 v) 
item it 
loquitur loqf 
Matheus mattie 
rneus ms (for 'meos' f. 71 r) 
raihi m 

modo mo (f. 229 r) 
(nobis nob 
\uobis uob, ub- (once) 
nomine nom 
non no, n 

^noster^ nf, nra, nfm, nri, nrae, nfo, 
nram, nrls, nros, nras 
(fully declined), nfm 'nos- 
nra 'nostram' (f. 206 v) 

na (na^), iim, nl, nae (f. 87 r 
corr. al. man. nfe), (ne), 
no, nam, nos, nas, nls 
nosra 'nostra' 
ns 'nostris' ((".^^ms^, f 117v) 
nit 'nostra,' nsm 'nostram' 

1 It is of very considerable interest to observe that iioster is frequently unabbre- 
viated in these phrases, dm nostri itu xpi, ihin xpm dnm nostrum. This fact sug- 
gests'that an ancestor of our MS had the early suspension h for any case of rioster. 

2 F. 130 V has na gloria na { = nam gloria nostra). Note the different positions of 
the abbreviation stroke. In this matter the MS is not consistent. 

3 Archetype must have had nis. 




uester ur, ura, urm, uri, uf §, iiram, 
ui'is, urs ' uestros '(f. 1 95 v. ), 
uras, Ufa 'uestram' (f. 
171r),uf'uestra'(f. 151v), 
ufs 'uestris' (corr. m2 
uns) (f. 254 v) 

ua, urn, uae, uam, uo, nils 

iiest 'uester' 

list 'uester,' 'uestra,' 'ues- 
trum,' usti-ii, 'uestrum,' 
ustra 'uestram' 

usrm 'uestrum,' usram 'ues- 

J mine nc, n 
\tunc t 

omnipotens omlTIps, omps 
omnis oiTis 'omnes,' 'omnis,' honis 
(f. 108 v) 
oma 'omnia' 

om 'omnes' (corr. al. man. 
f. 16 r), 'omuis,' 'omnem,' 
omns (hofans) 'omnes,' 'om- 
nis': omna 'omnia,' omn 
'omne' (f. 192 r) 
passione pass 

Pauli PAU, PAUL 

per p, p, p- (Caroline scribe twice) 
PIdlipenses phit, philip, philipens, 


populus ppls, pplu 'populum,' pplm 

'populum,' pplo 'popiilo' 

post p; p' (also in p'tulent 'postu- 

lent,' ap'tlo 'apostolo,' p'tea 


potest p (f. 235 v) 

prae p (puaricator, f. 112 v, must be 
due to confusion of p and p, 
because ji in this scribe 
means per) 
presbyter pfbi 'presbyteri,' pi-bni 
presbi ' presbyteri ' 
pribi 'presbyteri' 
prbris 'presbyteris' 
presbiX 'presbyterum' 

pp (pp, first stroke erased, 
f. 133 r), pp, pp 



publicamis puli- (f. 215 r) 
-gu- q^' 

qua q;_(f. 140y)i 
quadraginta XLta 
quae q; (sometimes corrected by 

another hand, as q; should 

be used for 'que' only) q; 

q:' a* q 
quaei'itur qr- 

qua7H ft 4 4 aH£t,ua (f. 220 r gives 
the 'nam' twice over, the 
second equivalent being an 
after-thought) g/ 

quaudo qiio, qn 

quare qre 

quasi qsi (qsi) 

que q;,ag'-»a 

qui q q q 

i a 

quia q< qa q q q ; 
quid qt 'quit' for 'quid' (three 

qms qs 

quod qd (qd qd) % qud qod (once 
qod) quod (once) ij^ q.(0 

* _ 
quomodo qmo 

quoniam qnm, qm (most character- 
istic), qum 
quoque qq 

reliqua reliq- rel; rel relq: rl 
respondit rF 
Romanos rom 
saecularis seclaria, solaria, 'saeou- 

saecidiim slo 'saeculo' (f. 60 r), secli 
'saeculi,' seclo 'saeculo, 
secla ' saecula,' seclorum 
sctm, scli, sola 
sanctificatio scificationis 'sanctifica- 

sanctitas scitate 'sanctitate' 
sanctxis etc. sea etc., scs 'sanctos' 
(f. 198 V his), sis (once) 

prima pri, prmi, p 
pr! 'primae' 

P^'o 9 , — _ - _ , 

propter ppt ppt, ppt, ppt, ppt, ppt, 

ppt, ppt 

1 A certain number of the 5 (/-abbreviations appear to have been ignorantly 





secunda secuTi, secund, seed, sec, 

secundum sec, seed (seed), seeund, 
secdu (seedu) 

semel semi (f. 254 v) 

seqtiitur seqT 

sequutUur seqnr 

siciU sic 

spiritalis spitalis: splritalia (f. 109 v) 

spiritus etc. sps, spni, spu, spui : 
spuu, spuum 'spiri- 
tuiim,' spurn 'spiri- 
tum ' 

Syllable Symbols : 

con c P (f. 140r) 

e d 'de' 

en m 'rnen,"niernbra'(f. 75r),mbra 
'ineml)ra' (the spelling menhra 
is fivquent in MSS) 

er t 'ter,' B 'ber' 

is b 'bis,' 1 'lis' 

it pel u b, incipt 'incipit,' cre- 
didt 'credidit' (f. 253 v) 

m suprascript stroke, generally of 
the collar-bone shape, slightly 
inclined from the vertical ^ (once 
at least beyond the letter in 
imitation of the ancient style, 

n suprascript stroke, as for m 

runt r 

V ' 2 

ur tr t tr 'tur' (never t)'^ 

&unt st (st), s (also posst 'possunt') 

tamen tii 

iantum tm 

tempore tempr (tempr) 

tertia ter 

Tesalonicenses tesol, thesal, tesalo- 
noc, tesalon, tesat, 

Timotheum timoth, timo 

\iel ul -u- (u) 


tiero u, uo 
uersus uer 
uicensima XXma 



; b 'bus,' d 'dus' (once, f. 85 v), 
1 1+ 'lus,' m; m^ 'mus,' n; n^ 
'nus,' p; «pus,' t 'tus' (?)(f. 
240 v), see eius^ 

if r}^ r (r~, see under m) r, 
(f. 160 v) 
nem u (fairly frequent), beatitudih 
'beatitudinem,' multitudiN 
'multitudinem,' etc. 
bun bprobabtur'probabuutur'(corr. 
al. man. probabutair), hab- 
dauit 'habundauit,' dicebt 
'dicebuntur' etc. 
ssime s- plenis- 'plenissime' (f. 220 r) 

(suchlike earlier also) 
centes et requiesct 'requiescent6s'(?) 
bin b cherub 'cherubin' (f. 251 r) 

The extraordinary variety of abbreviation here would seem to 
indicate that various influences were converging on the centre where 
this manuscript was produced, and that no standard set of abbrevia- 
tions had yet been evolved for that scriptorium. What may be 
regarded at once as certain, is that the immediate parent of this 
MS was in insular, in fact Irish, script. How otherwise could one 
explain the fact that insular abbreviations abound most in the part 
which is written in the Caroline script ? The following symbols are 
insular: h", c, dt, dnt, dnr, h, m, p*, qsi, g,, tn, tm, Z> (=con), and 
the following are definitely Irish rather than Anglo-Saxon: ap, dre, 
h, qn, qno, qre, qmo. We need not therefore doubt that the im- 

1 What Prof. R. S. Conway calls the 'signum Floriacense' : see bis note on 
Livy V 43, 1; also the preface to his vol. ii p. xxv. 

2 Dr Lowe tells me that t first appears about .k. d. 800, at Tours. 

=* From this point onwards the abbreviations do not figure in Prof. Lindsay s 


mediate ancestor was in Irish script. But this is by no means 
the only element. The following abbreviations are definitely non- 
insular: d, e, it, sic, ul, u-, f (=runt). Some of the symbols have 
a decidedl}^ Spanish look: such symbols for apostolus as show the 
stl, stls, sis; episcopm; nst, nsn5, ust, usrm etc.; omns, homiis etc. 
An earlier stratum appears in the suspension abbreviations apos, 
epis, h (= noster, which we have postulated) etc. 

The orthography is of the worst Merovingian t}^e : u and o are 
constantly interchanged, as in cognusco, subauditoi\ actos; short e 
IS substituted for short i, as in crededi etc.; the aspirate is often 
added, as in himagines; t is found for c, as in mendatium, and c for 
t, as in noticia; long i for long e as in adimplicio for adimpletio: 
short i for short e, as in habit; y for i as in cyhus; internal aspiration 
is found, as in typhus for typus; ae for e as in distruaerit: abstullit, 
necglegere, uellit, presura, abssens, alico, escandalizant, scilitit, stote 
etc. Certain of the abbreviations of proper names given above also 
show how freakish the orthography is on occasion. 

One or two notes with regard to the script may be here collected. 
Dr H. M. Bannister was struck by the forms of capital A R and Q. 
The R has the upper loop separate from the lower curve. As we 
find this also in certain Visigothic MSS, it has occurred to me that 
it may have been consciousl}^ imitated from the Visigothic ancestor 
already postulated. Bannister regarded the collar-bone suprascript 
stroke as indicative of an origin East of the Rhine, away from court 
influence, and he found many North German marks in the MS. 
The accents on monosyllables like es, o, os, a, ae, his are no doubt 
derived from the Irish parent. The letter u sometimes takes an 
extraordinary form after another u, the form being like a Greek 
minuscule stigma: us\t (f. 123 v), u^ltis (f 173 v), paruilus (f. 123 r), 
nomm (tf. 125 r, 170 v). The MS also uses a form like C to re- 
present V in numbers, thus G = VI (f. 238 v): the only parallels 
to this known to me are in MSS Gotha, membr. i n. 85 (from 
Murbach, saec. viii — ix) and Vat. pal. 574 (from Lorsch, saec. ix)^ 
Dr Lowe was struck by the resemblance between my rotographs 
and the minuscule part of MS Epinal 68 (saec. vii — viii) (formerly 
of Murbach). Prof Lehmann, after inspecting the same rotographs, 
declared for the border of France and Germany as the place of 

^ Das Decretum Gelasianum...^. v. Dobschiitz (Leipzig, 1912) p. 141. 


origin of the MS. Dr Alfred Holder recognised the Reichenau m, 
but to my great regret I never obtained the results of a detailed 
examination from him^ Lehmann, Lindsay and Lowe have all 
suggested Murbach in Alsace as the place of origin of the MS, but 
with a query. In the Notae Latinae^ the onus of this opinion is 
laid upon me: "according to Souter, the contents... suggest the 
possibility that it was written at Murbach." I will presently proceed 
to show how it is that the contents do suggest this, but first I ought 
to allude to two other MSS, which certainly belonged to Murbach, 
and which offer some analogies with M. They are Oxford, Bodl. 
Junius 25 (saec. Vlll) and Manchester, John Rylands Library 15 
(saec. viii). Both these MSS were written by a large number of 
scribes: so was our M. Such manuscripts are on the whole ex- 
ceptional, and it may have been a regular practice of the Murbach 
scriptorium at the end of the eighth century to apportion the task 
of copying among a large number. Again, a close study of the 
abbreviations in the Rylands MS shows many interesting points 
of contact with ours^ 

That there was a manuscript of Pelagius in the Murbach library 
at the middle of the ninth century, we learn from the catalogue of 
date about a.d. 840, published by H. Bloch^, where the following 
entry occurs: 

210. Exposicio Pelagii in epistolas Pauli. 
The catalogue, moreover, contains another entry only less interest- 
ing: it is this: 

39. Canones leronimi et Pellagii. 
Now it happens that our MS comprises the contents of both cod. 
210 and cod. 39, if the Canons in our MS may be identified with 
those in cod. 39, as surely they may be, seeing that internal evi- 
dence shows them to be of Pelagian origin^ But the Epinal MS 
which falls to be considered next, has the same contents, and it is 
possible that we ought to identify the Epinal MS with the Murbach 

1 My rotograpbs were in his possession from Sept. 1913 till his death in Jan. 1916. 

2 P. 471. 

^ See 'List of Abbreviations and Contractions, etc., in the John Eylands Latin 
Manuscript No. 15' by the present writer in Bulletin of the John Rylands Library, 
Manchester, vol. v (1918—1919) pp. 111—115. 

* Strassburger Festschrift zur 46. Philoloyen-Versammlung (1901) pp. 271, 276. 

5 See below for references to De Bruyne and White. 


MS. As a matter of fact, the Epinal MS was once in the Abbey 
of Moyenmoutier, but it may nevertheless have been earlier in 
Murbach, like Epinal 68 (saec. vii — viii) and Epinal 78 (saec. ix), 
both of which were in Moyenmoutier in Montfaucon's time, but 
are nevertheless Murbach books. If we identified our Epinal MS 
with the Murbach Pelagius, then it w^ould be tempting to identify 
the Paris MS (M) with the Lorsch Pelagius \ Yet even if M was 
once in Lorsch, it is not, Professor Lindsay tells me, a product of 
the Lorsch scriptorium. Murbach and Lorsch were as a matter of 
fact in close relations with one another-, and we should expect 
their Pelagius MSS to be related with one another. 

The MS *39. Canones leronimi et Pellagii,' which must have 
been a small MS, has probably perished. It is not impossible that 
it was an autograph, for the work is a pseudonymous production 
hardly likely to be earlier in origin than the seventh or eighth 
century. It is evidently based on the well known Canons so often 
found in Vulgate MSS of St Paul's Epistles^ But if it is lost, at 
least three copies which go back ultimately to the autograph, are 
still extant. The two copies discovered by me take precedence in 
age over the third copy, which is evidently that employed by 
Vezzosi for his supplement to Tommasi's works'*, and rediscovered 
and recollated by De Bru3rne at Gotha in the Biblical MS membr. 
I 20 fol. 217 (saec. x), w^hich is undoubtedly a Murbach book. 

Murbach was in fact a very important centre. It was founded 
from Reichenau by Pirmin in 725, and both Charlemagne and 
Alcuin were in touch with it. The territory of the abbey extended 
as far as Lucerne in the ninth century. It was somewhat ravaged 
by the Swedes in the sixteenth century, but recovered its glory 
in the seventeenth century, when it passed to France, Abbot 
Bartholomew of Andlau had ordered a catalogue to be made in 
1464, which catalogue exists both in Latin and in a French trans- 

1 See p. 28 above. 

2 Hauck, Kirchewjeschichte Deutschlands Bd. ii pp. 593 f.; Von Dobschiitz, Das 
Decretum Gelasiaiium p. 143. 

•' 'Una Concordance Biblique d'Origine Pelagienne ' in Revue Bihlique, t. v (1908) 
pp. 75 — 83; Wordsworth and White, Epistula ad Romanos (Oxon. 1913) p. 12. I 
recollated the Canons myself at Gotha in 1913 and the Vezzosi text, a scarce book, 

^ Thomasii... Opera Omnia, torn, i (Romae 1747) pp. 489 ff. 


lation^ In that catalogue the Canons MS still appears: 'n. 40 
Canons de Saint Jerome et de Pelage-,' but of the manuscript 
of the Pelagius commentary there is not a trace. About the 
seventeenth century the Murbach library was transported to 
Guebviller, and then by the Republic in 1791 to Colmar. There 
and earlier considerable leakages have taken place*, and only fifty 
or sixty volumes are now known to exist, of which thirty-four are 
at Colmar*. One of these is a MS of the Pauline Epistles (saec. 
viii), which does not contain our Canons, and does not seem to 
agree in Biblical text with my M or N^ The connexion of our M 
with Murbach is, therefore, not absolutely certain. 

(8) Epinal, No. 6 {saec. ix in.) (N) 

This manuscript belonged to the Abbey of Moyenmoutier" in 
the neighbourhood: fol. 1 r top reads 'Mediani Monasterii Catalogo 
inscriptus 1717.' Old shelf-marks are: 

X 1 ) 

,, ^„ y Arm- No. 6 45 49 (these two are later than the 
No. 23 J ^ 

others). All these are preserved on the cover, which is not older 

than the sixteenth century. 

The MS now consists of 176 ff. (unnumbered). As the codex 

breaks off in the midst of a comment on section xxiiii of the Epistle 

1 Lettres et Pieces rares ou inedites publiees et accompagnees cf introductions et 
de notes par M. Matter (Paris, 1846) pp. 40—76. 

2 P. 46. 

3 See the present writer in Bulletin of the John Rylands Library, Manchester 
vol. V (1918— 1919) pp. 392 f. 

^ P. Lehmann, lohannes Sichardns . . .CMixnchen, 1911) pp. 164 — 175. 

5 I have been favoured with select readings by Dr White, who now possesses 
a photograph. Literature on Murbach will be found in Lehmann, loc. cit. I have 
to thank Monsieur Leon Dorez and Doni Wilmart for references to A. Gatrio, Die 
Abtei Murbach in Elsass, 2 Bde (Strassb. 1895); F. W. E. Roth in Strassburger 
Studien Bd. in (1888) p. 339; Montfaucon (from Calmet), Bibliotheca Bibliotheca- 
rum...ii (Paris, 1739) pp. 1175 — 1178; Mabillon and Ruinart, Voyage Litteraire de 
deux religieux Benedictins (Paris, 1717) ii p. 138. 

s On Moyenraoutier, Dom Wilmait kindly refers me to Montfaucon, Biblioth. 
Biblioth. II 1180, 1759 (cf. 1175); Calmet, Ilistoire de Lorraine, t. vii (2) (1757) 
pp. cii — civ; M. Jerome, Histoire de I'Abbaye de Moyenmoutier, t. i (1902) (down to 
sixteenth centm'y); Th. Gottlieb, Ueber mittelalterliche Bibliotheken (Leipzig, 1890). 


to the Hebrews^ the portion lost, assuming that it was identical 
with that in M, where it takes up six and a half pages, was not 
sufficient to fill a whole quaternion. The Canons at the beginnings 
(occupying the first quaternion) are arranged in two columns' to- 
the page, but elsewhere there is one column only to the page. 
Each page contains 33 lines^ and measures 335 mm. by 218 mm.^ 
while the written part measures 290 mm. by 172 mm. The MS 
appears to have been executed by at least three scribes. The first 
wrote quaternions 1 — 11 inclusive (=fi". Ir — 88 v): the second 
wrote quaternions 12 — 20 (= fif. 89 r — 160 v) inclusive (except a 
small portion of fol. 117 v): the third wrote the first eleven and 
a half lines of fol. 117 v, and also quaternions 21 and 22 (= ff. 161 r 
to 176 v): a fourth seems to have written the last four lines of 
f. 136 r, perhaps most off 149 v, all off 150 r and the first three lines 
of f. 152 r: traces of other scribes are found here and there. The 
twenty-two quaternions which now compose the MS are all abso- 
lutely regular: .!• signed below the right-hand column of f. 8 v: | II [ 

(below, middle of f. 16 v): so with _m! IIII \V_\ \Y1\ VII VIII : 

the ninth, tenth and eleventh bear no numbers; the 12th to the 21st 
inclusive (which, as we have seen, are by the second scribe) are 
lettered below the text, in the middle, on the respective pages, 
A B C • ^ ■ -e- F" Q" • h • I • K ■ (for the most part in uncial letters). 
The twenty-second and last (surviving) quaternion is neither 
numbered nor lettered. The vellum is mainly thick and strong. 

The contents are described above, but the commentary on 
Hebrews does not appear to agree closely with that in M, which 
suggests that while for the thirteen Epistles the two MSS are 
closely related, they derived their Hebrews expositions from 
separate sources. 

At the foot of fol. 1 r the same hand (probably) which wrote 
the note of Moyenmoutier ownership, has written: 

'D. Hieronhni seu potius Pelagii heraesiarchae Commentarius 
in epistolas Pauli! 
The scriptural lemmata are underlined with red ink down to 

1 At chap, vii 24: see Riggenbach, Die dltesten lateinischen Kommentare zum 
Hehrderhrief (=Zahn's Forschungen u.s.w. 8 Teil) (Leipzig, 1907) pp. 205 fif. 

- But the first scribe has crushed 35 Unes of small writing on to fol. 88 v in 
order to finish his portion there. 


1 Cor. xiii 9 et ex parte prophetamus (f. 75 v). The characters are 
large and well-formed, and belong to the type associated with 
Murbach and related centres^: the second scribe in fact recalls 
the work of the first scribe of A, but he is hardly so elegant. 

Quite early the MS was injured by damp and mice(?), as 
illegible words were supplied even before the rubricator began his 
work, that is, at earliest in the middle of the thirteenth century-. 
The damage affected the tops of the leaves, which are partly dis- 
coloured. The codex is too lavish with punctuation, by means of 
a dot placed after every two or three words. These points, the 
work of the original scribes, have been, for the most part, scraped 
out. This gives the MS a worse appearance than it would have 
had, if they had been suffered to remain. Portions of the margins 
of folia 78, 79, 92, 95, 98, 105, 107 and 127 have been clipped away, 
and a rent in fol. 91 has been most skilfully repaired by sewing. 
The first scribe gives no headings to the pages, but the second 
scribe, for example, heads f 89 r (his first page) thus: 


The outer pages of folia are regularly the smooth side of the 
vellum, not the hair side. The following trifles occur at the points 
indicated : 

mg. f 102 V Virtus \ lanipat | d- sotiat • | [N]atiuitas | unde 
(saec. XIII?). 

mg. f. 103 V The alphabet is run up the margin, A being given 
in uncial and minuscule, while the other letters are in minuscule 
only (once each). 

mg. f. 106 V Virtus lampat & social Natiuitas unde nobis (same 
hand as on f 102 v). 

mg. f. 146 v klmnopqrs (rest cut off by binder: a probatio 

mg. f 147 V ahcdefg (rest cut off by binder: a probatio 

As regards the forms of letters, etc. it may be remarked that 
open a is very common, that half-uncial F is sometimes found, and 
that the second scribe occasionally uses the capital R and S in the 
middle of writing otherwise minuscule. Accents are occasionally 

1 See p. 204. - Judging by the characters on f. 72 r. 

S. P. 20 


found on the vowels of monosyllables, insular fashion: 6, ef. The 
third scribe has a curious habit that I do not remember to have 
observed elsewhere, the use of a circumflex accent over the first 
consonant, where that consonant is doubled: thus accione , accipere 
(e.g. f 167 v). The mark has doubtless something to do with pro- 
nunciation. Omission from homoeoteleuton, afterwards corrected, 
is quite frequent. 

With regard to orthography it is enough at this stage to men- 
tion that aspirates are often wrongly present, or wrongly absent, 
that we also often find a consonant single, where it should be 
double, and that the second scribe spells secondum, iracondus. 

The abbreviations in this manuscript are not nearly as numerous 
as those in M, but they are of considerable interest. It looks as if 
the scribes of N had received instructions to use as few abbrevia- 
tions as possible, and it is therefore not improbable that most of 
those actually employed are taken over from the exemplar. I have 
added the letter M in brackets after each abbreviation of N which 
we have found also in M. The evidence will show, I think, that the 
two MSS belong to the same region, even if it does not prove 
that all abbreviations found in both belong to their common 

apostolus apostl (nom.) (M), apostl dicit dit (f. 64 v) (M) 

(ace. sing.), apostl (ace. dicitur df (M) 

pi.) dominus etc. dnsetc. (M),dmn(?)'do. 
apost (M) miniim' (once) 

apl (M) eius eis (ff 27 r, 3.3 r) 
aps (M) ei; (third scribe only, several 

apols times) (M) 

apostols enim -H- (M) 

apostul episcopus epos 'episcopos' (M) 

autem au (first and second scribes) epistula epis (M), ep 'epistulae' (M), 
(M) epi 'epistulae,' 'epistii- 

aut (first scribe rarely, third larum' (cf. M) 

scribe always) (M) e««« ee (M) 

aum (m eras., f. 26 r)_ «««e< eet (M) 

capitulum cp 'capitula,' cap 'capi- ^^^ -^ (M), e (M) ,. , ,« 

tula' (M) euangelium guanglo euangelio (f. 

carissimus kms (M) . . _ |"'* ^) 

n- expositio exp (M) 

cetera c&v; __ .^^^^ fr.TS (f. 136 v) (M) 

C%r,.;tan./^_xpianus (M) ^^^ ,^^.^^g^, ^^^^ ^^ ^^ ^^^ 

CAm^t.. xpsetc (M) gratia gv^ 

n . ... ^P^"^'V^"«t;«n (f-36r) ^Hiermob^ma hierusol 'hierusolymis' 
Connthu chrs .'corinthic^ cormt ^^^^^^ ihus (f. 72 r) 

dms etc. dsetcTM)'""' ^'"^P^ inept, incip (M) 




lohannes ioh (M) 

iotlH 'lohannis' (f. 36 r) 
Israhel isrhl (M) 
mens mi (f. 58 v) (M) 
mihi mifi (third scribe, thrice) * 

in (f. 168 V his) 
(nobis nb, noB (M) 
\uohis ub 
non n (M), no (M) 
(iwster'^ nt (M), na (M), nm (M), 
ni (M), nae (M), no (M), 
nam (M), nis (M), nos 
_(M), na5 (M) 
nr (M), nfo (M), nram (M], 
nfm (M), nfa (M), nri 
(M)(thistype third scribe 
^' always) 

nrt (f. 141 r) 
nost (M) 
uester ut, ua (M), um (M), uae (M), 
ur (f. 135 r) (^I) 
usta (corr. ustu) 'uestrum' 
\^ (once) (cf. M) 

omnipotens omps (M) 
omnis oms 'omnes' (M) 

om (for various cases) (M) 
oniia (first scribe), oma (se- 
cond scribe) (M) 
Paulus pauls 
per p (M) 
post p; (M) 
prae p (M) 

presbyter prbi 'presbyteri' (f. 161 r) 
(M), prbim 'presbyte- 
rum' (f. 161 r) 
pro <p (M) 
propter ppter (M) 

ppt (M), (ppt [M]) 

Syllable Symbols : 

con c (M) 

en iTi 'men' (M) 

er b 'ber' (f. 171 r) (M), t 'ter' (M) 

is b ' bis ' (third scribe) (M), d ' dis ' 

it c 'cit' (M), p 'pit' (M) 

PP (M) . 

prop (corr. propt) (once) (M) 
<j)pte (e eras.) (once) 
quae q: 

quasi qsi (f. 172 r) (M) 
que q(M), q;(M) 
quia Cfi (cf. M) 
quid qt (from 'quit') (M) 

quod qd (M) 

quomodo quomod (f. 16 v, end of 
quomd (third scribe, thrice) 
quodo (once)^ 
quo7iiam* qUo (frequent) 

qm (sometimes) (M) 
quom (f. 24 r) 
qiim (twice) (M) 
reliqua reliq, rliq, relq (M), reli, rel 

(M), rlq 
Romanos rom (M) 
saeculum secli 'saeculi' (f. 162 v) 
(M), seclof 'saeculorum' 
(f. 168 v) (cf. M) 
sclo 'saeculo' (f. 162 v) 
(_cf. M) 
sanctus etc. scs etc. (M) 
sicut sc (f. 162 v) 

sic (twice at least, once where 
true text is si cum) (M) 
spiritalis spitalis (M) 
spiritus etc. spi etc. (M) 

spiim 'spiritum'(f. 55 v) 
_ (M) 
sunt s (M), St (f. 160 r) (M) _ 
Thesalonicenses thesaloncenss (once) 
uel 1, ul(M), u(f. 144r), (M) 
uero uo (M) 
uersus uer (M) 

m suprascript stroke (M) 
n suprascript stroke (M) 
rum r^ (M) 
runt f (M) 

us ts, ts 'tus' 

1 Lindsay, Notae Latinae, p. 125, explains this as a scribe's alteration of mfi of 
his original. This would suggest that the original was Veronese, but I know no 
other indication in N pointing to such ancestry. 

2 Written in full, ff. 27 v, 143 r. 

■^ This and some other abbreviations were by oversight not communicated to 
Prof. Lindsay. 

■• Abbreviated apparently by the first scribe only. 



There are some notable features in the abbreviations that N 
does not share with M : apols occurs in a Fulda MS and in one now 
at Troyes; aum clearly indicates a Spanish stage in the trans- 
mission^; eis appears to be unknown elsewhere; mih and m; nb 
and ub, doubtless copied straight from an insular copy; nrt appears 
to be not older than a.d. 800; traces of quomod and quomd are 
found at Flavigny (Autun) and Peronne respectively; all our ab- 
breviations for quoniam might occur simultaneously in an insular 
original; sc is very important, as it occurs in early specimens of 
insular, and is no doubt taken straight from the original^; the same 
may doubtless be said of u, shared with M'*, 

Here also then we seem to find clear proof that the immediate 
original of N was in insular script, and that there was behind this 
insular stage a Spanish stage in the transmission. And the argu- 
ments for a Spanish stage in the transmission of Hg are not ex- 
clusively palaeographical*. In tJie cases both of M and N then we 
have arrived at this conclusion: 

Spanish MS 

Spanish MS 

(saec. VII — viii) 


(saec. vii — viii) 


Insular MS 

Insular MS 

(saec. viii) 

(saec. viii) 





It is time to ask what is the precise relationship between 
M and N. 

Before discussing whether the one MS is a copy of the other 
or not, it would be necessary to know for certain which was the 
older. M has altogether a more antique look, and probably most 
palaeographers would agree that it is rather older than N. The 
close relationship of the two MSS is not in fact a mere matter of 
sharing abbreviation symbols. They also share thousands of wrong 
readings, as well as the set of canons and other prefatory matter 
to which allusion has already been made, and the numerous inter- 
polations which I hope to edit in a third volume. A few illus- 
trations of wrong readings, where other MSS are right, may now 
be given. 

1 Lindsay, Notae Latinae, p. 25; C. U. Clark, Collectanea Hispauica, p. 84. 
- Lindsay, Notae Latinae, pp. 286 f. ^ Lindsay, Notae Latinae, pp. 311 f. 

■* See pp. 271 f. above. 


MN True Text 

In Rom. V 9 ciistodiat custodiet 

20 dicerint dicerent 

vi 6 noster^ nostrum 

vii 13 (interp.) ineipit incipiat 

M cannot have been copied from N, because at the famous 
lacuna in Rom. v N has no gap, while M has, as we have seen, a 
most precisely measured gap I If M had been copied from N, it 
would have gone on without interruption, as N does. It is possible, 
if not probable, also, that if N had been copied from M, N also 
would have represented a gap, even if not so striking as that in Ml 
But there are hundreds, if not thousands, of differences between 
the readings of M and N which show that neither can be a copy 
of the other. 

In Rom. 



True Text 

ii 4 



as M 

V 14 




vii 13 

bonam legem 

bonam legi 


viii 3 

(interp.) quoniam 


as M 

viiii 26 



as M 

viiii 33 

Petrus quoque — con- 


as N 

fundetur o»i. homoeot. 



signal proof) 

xi 1 


consolatur {ex 


as N 

. vi 15 

id est — meretricis 


om. homoeot. 

as M 



signal proof) 

1 Cor. 

Such differences are in fact so very numerous and serious that 
it is inconceivable that M and N are both direct copies of the same 
MS. Between each of them and their common original there must 
have been at least one codex interposed. No doubt both these codices 
were insular, but I think M was copied direct from an Irish MS, 
and N direct from an Anglo-Saxon MS. Not one of the definitely 
Irish abbreviations which we have cited from M*, occurs in N, and 
on the other hand N nearly everywhere employs quo, which is an 
Anglo-Saxon and not an Irish symbol. This Anglo-Saxon symbol 

1 Due no doubt to a wrongly expanded n, which does duty for every case. 

2 Cf. p. 271, 

* Cf, the parallel case of B and the Merton MS, pp. 216 f., 225. 

* The same corruption occurs near the beginning in another interpolated part, 
due to wrong expansion of quo, 

5 Above, p. 299. 


never occurs in M, which habitually employs the Irish symbol qm. 
Furthermore, as will appear immediately, M often agrees with G 
(which we have seen was copied from an Irish exemplar), against 
N and the corrector of R. Now, we should expect an insular MS 
accessible to a Ratisbon scribe to be Anglo-Saxon rather than Irish. 
I should therefore sketch the connexion between M and N thus: 

Spanish (Visigothic) (saec. vn — viii) 

Insular (saec. viii) 

Irish (saec. vni) O O Anglo-Saxon (saec. vni) 

M N 

(9) The lost MS used by the corrector o/R 

One passage will serve to show what the connexion between 
the corrector of R and M — N really is. Let us take an interpolated 
passage, for which we have also the authority of G. It will be con- 
venient to print the text of M and to record the variants of the 
other manuscripts in an apparatus. The passage comes after the 
word 'domini' {in 1 Thess. iiii 15) in MN, after 'inueniret' in R corr. 

Hoc loquo hostendit q. in die iudicii sub aduentu diii quicumque sanctorum 
in ac uita inuenti fuerint non prius aduentante diio obiam rapientur in nubibus 
caeli nisi sci omnes qui superioribus retro temporibus in Christo dormierunt 
resurrexerint et tunc demum omnes simul hoccurrere diio in aera in nubibus 
5 adsumentur nequis autem ambigat et extimet scs in aduentum diTi gustaturos 
mortem nee post aliquem hominem natum non mortem desoluet hoc dr ad- 
firmante apostulo paulo q; in morte xpi baptizati sumus et consepulti cum 
ipso in baptismo in mortem. 

1 loquo] loco NRG host.] ost. NRG q] quia NRG aduentu 

{'eras.) N quiqumque N* 2 hac NRG adueniente NG obuiam 
NRG 3 caeli] celi N om. RG sancti om. R homnes (h eras.) N qui] 
-Hin G superioribus om. NR temporibus] in tempore NR co dor- 

mierunt in Christo NR 4 occurrere NR occurri G 5 et extimet] et 

exestimet G om. NR aci ( = sanctus)] scos NRG aduentu RG diio N 
6 post] potest NR posse G hominum G morte G desoluet] desolue N 
deaoluere R desolui G dicitur {in full) G 7 apostolo RG paulo om. 
NRG q;] quia NRG babtizati R 8 illo G babtismo R baptisma G 
morte R. 


The freakish (some of them Spanish) spellings of M are not 
shared by other MSS. There is a real relationship between N and 
R, which is not shared by M. M goes rather with G, though G is 
as elsewhere somewhat independent. Not infrequently M takes its 
own line against all other representatives of Hg. 

The copy used by the corrector of R appears to have been a 
sister-MS of N or of the immediate predecessor of N. The relation- 
ship between N and R corr. will become clearer when the inter- 
polations in Hg are critically edited. It is impossible to suppose 
that R corr. had any surviving MS in front of him. 

(10) Troyes, 486 {saec. xii^) (C) 

Signs of former ownership and older shelfmarks are preserved 
as follows: (f 183 v, foot): 'Liber See Marie clare uallif (i.e. the 
great Cistercian Abbey of Clairvaux, immortalised through its con- 
nexion with St Bernard): (f. 184 r after the end of the MS) 'f 43 
(43 erased) QQ,' and below this 'f. 66.' These are the shelf-marks 
in the Clairvaux library, as is confirmed, if confirmation were needed, 
by the evidence of the ancient catalogue of the fifteenth century^, 
of which a copy made in 1645 exists at Paris ^. The title of this 
copy is: 'Inventaire, et Declaration Des Volumes, et Livres de 
L'Esglise et Abbaye de Cleruaux De L'ordre de Cisteaiix oii 
Diocese de L'Engres Fait ou Mois de May, L'An Mil Quatre Cens, 
Soixante et douze Par Nous Frere Pierre Nouuel Abbe dud. Lieu. . . 
Ledit Inuentaire Escript de Nouueau, En Januier, L'An Mil six 
cens Quarente cinq A Dijon MS De la Bibliotheque de M'' le... 
Bouhier A. 52 mdccxxi.' 

On f 39 the following entry occurs, referring to our MS: 
' Item Un autre beau Volume conten. Explanation s* leome . sur 
les xiiij Epres S'^ pol. qui e nome, breuiariu. Sci leonlmi Com- 

^ M. L^on Dorez says not later than the middle of the century ; Dr Bannister 
assigned it to the end of the century or to saec. xiii. (This latter date must be 
rejected in view of Dom Wilmart's discovery; see below.) I am deeply indebted to 
M. Dorez for much valuable help in connexion with this MS. He has made a special 
study of the MSS at Troyes. See also Dom Wilmart's exquisite tractate, L'Ancienne 
Bibliotheque de Clairvaux (Troyes, 1918 [dated 1917]) from Memoires de la Societe 
Jcademique de VAube, t. lxxxi [191/]). 

2 Troyes MS 521. s g. n, pr. 22,364, 


mencat ou V Feuille qd pdixit ysaias. & Finissat ou penult entie. 
escript manentes ho. Sig' F. 66. 

Dom Wilmart has discovered in a gnard-leaf of Troyes MS 
32 (of the end of the twelfth century) a portion of a still earlier 
Clairvaux catalogue, which belongs in fact to the twelfth century. 
Our MS fortunately appears in it under libri sancti ieronimi (88): 

Breuiarium eiusdem super omnes epistolas pauli in uno iiol} 
Wilmart rightly identifies that MS with the surviving F 66: 
Troyes 486. 

The twelfth century is the Golden Age of the Clairvaux library. 
At that time, no doubt under the influence of Bernard himself, 
whose passion for learning seems to have equalled his holiness 
and his wonderful administrative ability, a large number of texts, 
some of them rare, were copied from manuscripts in every possible 
quarter. It is known, for example, that Spain and the Rhine country 
were laid under contribution. Dom Wilmart has recently discovered 
a Clairvaux MS at Troyes (no. 523, saec. xii), which contains the 
only known Latin rendering of a large number of discourses of 
Eusebius of Emesa, as well as five treatises of TertuUian and an 
opuscule of Pontius Maximus^ Nor are these texts merely careful 
copies of earlier MSS. They are corrected with a learning and 
intelligence unique in my experience'. Whether Bernard himself 
or some notarius of his was the corrector, I do not know. 

The Pseudo- Jerome MS has had its edges clipped all round 
and ' marbled ' by an eighteenth century binder. The pages now 
measure 325 mm. x 225 mm., each bearing two columns of writing 
measuring 240 mm. x 75 mm. The MS contains 184 folia. Some 
of the quaternion numbers have disappeared through the action 
of the binder, but it is possible to say that the manuscript consists 

1 Wilmart, op. cit. pp. 30, 32. 

2 Wilmart, op. cit. pp. 39, 43; Journal of Theological Studies, vol. xix (1917—18) 
pp. 316 f.; Analecta Bollandiana, t. xxxviii (1920) pp. 241—284; Academic des 
Inscriptions et Belles-Lettres, Comptes Reudus des Seances de I'Annee 1920 pp. 
380 ff,; Revue de I'Orient Chretien, t. xxii (1920—21) pp. 72—94. I have to thank 
him for copies of these articles. See also A. Souter, TertuUian Concerning the 
Resurrection of the Flesh (London, 1922) pp. 162—196. I have collated all the five 
treatises of TertuUian. There is the same careful correction in this MS that we 
have found in our C. 

"• I infer that the hundreds of other MSS are like the two I have studied. 


of twenty-three quaternions. The guard-leaf (fol. 1) is separate, 
and the signatures now traceable are found thus: I (f. 9 v), III 
(f. 25 v), nil (f. 33 v), V (f. 41 v)...IXi (f. 73 v)... (wanting f. 105 v) 
...XIX (f. 153 v)...XX (f. 177 v). The last is thus constituted: 

178 179 180 181 I 182 183 184 x 

It is possible that the MS is not now absolutely complete. 
The Canons which we have found in M and N, would have filled 
a quaternion, and it is conceivable that they once formed a part of 
the MS, though not quaternionised. On the other hand Bernard 
may have objected to them as heretical in origin, and ordered them 
to be suppressed. There is of course the further possibility that 
they were not in the archetype of the Troyes MS at all. The rest 
of the prefatory matter appears exactly in the same order as in the 
older members of the family, not only at the beginning of the codex, 
but at the commencement of each Epistle. It contains the same 
Hebrews commentary as that found in N^ We have already seen 
that M differs somewhat from N in that exposition. 

The following notes indicate the learning and care bestowed 
upon the text: 

(fol. 28 va mg.)j 'Nota lector quod fere ubique textus et expositio trans- 
posita sint et sibi inuicem confuse mixta, ita ut praecedat expositio et sequatur 
textus, quod nisi diligenter aduerteris, totum sensum confundet.' 

(fol. 30 b) after a sign : 'Quos autem p(rae)d(estinauit) requiritur ad prae- 
missum signum uerso folio. Hoc secundum signa post paginam ab (sign) usque 
sacerdos.' On fol. 30 va we find the same sign (like a capital L, somewhat 
ornamented in its transverse stroke). 

(fol. 91a mg.): 'capitulum alterixis loci secundum notulam superpositam ' 
(section viii of 2 Cor.). At end of section ix above 'quod' has a sign with 'ipc,' 
and in the margin opposite the same sign with 'capitulum alterius loci iuxta 
indicem superpositum.' 

Similar marginal notes occur ff. 94 vb, 106 b, 119 b. 
The corrector altered defective texts of scripture from a copy 
in his hands. The sign of this is crushed writing in rasura. 

1 Note IX, not vim. Whether anyone has discussed the age of the symbol ix in 
MSS, I do not know. 

^ Riggenbach, Die dltesten lat. Komm. zuni Hebrderbrief, p. 206. 


After the colophon (f. 183 v red) 'Explicit sci ieronimi expositio 
in Xlill eplis pauli apli' follow: 

'Versus d, ieronimo editi ad damasum papam. I(rec?)am dudum 
saulus...(f. 184 ra) monstrare triumphos. {red) Expliciunt.' F. 184 
is blank. There are many lovely coloured initials in the book, some 
of which recalled to Dr Bannister the work of Limoges and Toulouse. 
Some rare colours are employed on occasion. 

The most interesting feature of this MS from a textual point 
of view is the treatment of the lacunae in Romans v. We saw that 
there were two blank spaces there in M, while in N the text runs 
on in both cases almost without warning that anything is wrong. 
In C however^ an attempt has been made to fill up the gap, and 
it is of great interest to observe that it was not Ambrosiaster that 
was used for this purpose, but the Cassiodorian (Pseudo-Primasius) 
commentary, and the procedure followed was openly avowed. After 
'praeceptum' (fol. 20 a = Rom. v 14) there is a gap of about nine 
or ten letters in length: then a new line begins with 'Item mors' 
etc., but before this, in the gap, a sign occurs, and there is another 
on fol. 20 b mg. (after 'grauia'). Opposite the first sign in the 
margin occur these words: 'Quantum signis distiguitur in exem- 
plari non inueni.' The words he did not find in the 'exemplar' are 

Item : Mors ex originali malo ueniens usque ad legem sola regnauit. Ideo 
dixit regnauit quia totum muudum generaliter occupauerat. Alia uero peccata 
nequaquam regnare uidebantur quia uon per uniuersos dominabantur. Natu- 
ralis enim iustitia uigebat in plurimos {corr. is) et cgtera delicta non hereditaria 
sed uoluntaria erant. Ergo obligatio primi hominis sola usque ad moysen per 
so mortem operata regnauit. A legis autem tempore genus aliud delictoruni 
in reatum neglectae legis accessit, quod ante eius promulgationem non poterat 
inputari. Ac sic introeuntibus nouis praeceptis iniquitas multiplicata est. 
Lex enim non ut toUeret peccatum, sed ut demonstraret uel uendicaret aduenit, 
et ideo hie abundantiam gratiae apostoli doctrina commendat, quia non tantum 
ad§ debitum quod solum usque ad legem uelud tyrannica dominatione regna- 
uerat sed etiam innumeras offensas praeuaricationis et reprobe conuersationis 
bonitas redemptoris credentibus redonauit. Propterea dixit indicium quidem 
ex uno id est delicto in condempnationem, gratia autem ex multis delictis in 
iustificationem. Ac sic prima obligatio ueteris oftensg usque ad legem sola 
dominata est, alia uero delicta usque ad interdictum legis non iiidebantur, uel 
non intelligebantur esse tarn grauia. 

1 And in its descendant F; of. also p. 317. 


Except for the last sentence, this is taken with slight modifica- 
tions from Cassiodorus (Pseudo-Primasius, Migne, P. L. LXVIII 
p. 440 D — p. 441 B). 

At fol. 21 r a (= Rom. v 20) Sicut ait saluator ■ cui plus dimit \ 
titur the sign is placed after dimit., and in the margin opposite we 
have the words: 'Nee hoc in gxemplari habetur.' After gratiae 
magnitudo et the MS proceeds thus: 

per indulgentiam remissionis, cresceret debitum caritatis. Manifestantur 
beneficia, cum e contrario indignorum meritis conparantur. Mors enini per 
Adam regnauit, uita per Christum. Regnauit inquit mors ab adam, usque ad 
moysen. Quod etiam sic intelligendum est, usque ad moysen, id est usque ad 
finem legis et initium gratiae debitum naturale regnauit. Sed cursum fenoris 
sui perdidit, post quam Christi sanguis cyrografum originale deleuit. Sed dicit 
aliquis. Ecce per cathecuminos et gentiles diuersasque nationes originis malum 
regnat. Quibus respondendum est. Tunc uere pestis late diffusa regnauit, 
quando medicus deerat, quando mortalitas generalis omnes ad inferna mittebat. 
Ubi uero portam paradisi redemptio de caelis missa patefecit, ubi mundo 
attulit uitae auctor salutem, mors jjerdidit potestatem, quae oblatis a salua- 
tore remediis, non perdit nisi uolentes. Ac sic iam non regnat quia regnum 
eius gratia regnante destructum est. 

This is for the most part taken from Cassiodorus (Pseudo- 
Primasius, Migne P. L. Lxviii p. 441 b — c), but there has been 
rather more editing here. 

The companion mark is after 'destructum est' (£ 21 a). After 
'caritatis' (near the beginning) there is a gap of a line and a half 
After 'destructum est' two and a half lines are blank. 

As to the relationship of this MS to the others of the Ha family, 
it has already been pointed out that it is closer to N than to M. 
If we take the passages selected above to show the relationship 
between M and N\ we observe the value of C without much diffi- 
culty. Riggenbach noted that in the Hebrews exposition it is a 
better MS than N. In Rom. v 9 it has custodiet rightly (MN cus- 
todiat); v 20 dicerent rightly (MN dicerint); vi 6 nostrum rightly 
(MN noster); vii 13 incipiat rightly (MN incipit). Again, in the 
other list, we observe the following readings: in Rom. ii 4 homini- 
hus rightly (N homines); v 14 iniustum rightly (M iiisto, N iniusto); 
vii 13 bonam legem rightly (N bonam legi); viii 3 (Interpol.) quoniam 
rightly (N quod); viiii 26 plebs rightly (N pies); viiii 33 has (with 

1 See pp. 308 ff. 


N) the passage wrongly omitted by M; xi 1 consolatur rightly 
(M auadliatur); in 1 Cor. vi 15 has (with M) the passage wrongly 
omitted by N. This character is borne by the MS throughout. 

The MS, however, is not faultless. There is a case of omission 
by homoeoteleuton, for example, at in Rom. viii 3 (interpol.), where 
it omits 'Dicens in similitudinem carnis peccati,' which M and N 

It can be proved that C also harks back to an insular exemplar. 
At Rom. xi 4 where MN have rightly ds, C* has dicitur which could 
only have come from dr, the insular symbol for dicitur, which owing 
to the similarity between r and s in that script, may be confused 
with the other symbol. Again, at ] Thess. v 23 (interpol.) quia was 
written, and afterwards corrected to quam, because the scribe at 
first mistook^ for g,. This confusion was possible in insular script 
alone. MS C will be descended either directly, or with one remove, 
from the insular MS (saec. viii) which we saw reason to suppose 
lies behind the parents of M and N. If it had come direct from 
one of these parents, it could not have been so free from error. 
Its immediate parent need not however have been in insular script, 
though it doubtless retained insular symptoms. In endeavouring 
to guess the locale of this parent, the analogous case of another 
Troyes (Clairvaux) MS may help us. Koetzschau in his edition of 
Origen-Rufinus De Principiis (Leipzig, 1913)^ appears to prove 
that the Troyes (Clairvaux) MS of that work (saec. xil, double 
columns, like ours) is a copy of the still preserved Metz (S. Ar- 
nulphus) MS (saec. x) of that work. The parent of our MS was 
probably found somewhere in the Rhine country, not too far from 
M and N; perhaps at Lorschl 

(11) Florence, R. Biblioteca Mediceo-Laurenziana, 
Plut. XV Dext. Cod. i {saec. xii — xiii) (F) 

This manuscript, formerly of Santa Croce, though it is the best 
of all the Pseudo-Jerome MSS, need not detain us long, as it is 
undoubtedly descended direct from C, with one codex only inter- 
vening. In proof of this, it is enough to mention the identity of 

1 Pp. xxxvii f. 2 cf. pp. 28, 302. 


contents, and the fact that the corrections of C find their place in 
the text of this MS. That F is not a direct copy, however, of C, is 
proved by the fact that in a certain number of places it bears signs 
of an intervening exemplar, into which some further corrections 
had been entered : for example 

Fn Rom. xv 25 ut ueniam deinceps F: deinceps ut ueniam C* 
cet.^; ut deinceps ueniam C corr. 

In 1 Cor. i 2 deus ipse F: ipse C cet. 

It would be easy to fill pages with evidence of the character 
of F, which I collated with the same fulness as the MSS already 
discussed, in order to be quite certain about its place in the genea- 
logical scheme. The MS is at points fairly heavily annotated in the 
margin in a fifteenth century hand by some reader who was in- 
tensely interested in the subject-matter. These notes have, however, 
no value for the restoration of the Pseudo- Jerome text, and I have 
ignored them. Another sign of the care with which the MS was 
studied may be seen in the marginal r{equiras), where text or 
teaching was of doubtful character. 

(12) Cambridge, University Library, Ff. 4. 31 {saec. xv) 

This manuscript, which was written by one W. More, connected 
with Oxford", is of the same branch of the Ho family as C and F. 
It is headed: 'Incipiunt breues annotaciones beati ieronimi presbiteri 
super ep{isto)lam ad romanos.' W. More was a scholarly man, as 
we see from his insertion of the passage from the 8th chapter of 
Cassiodorus (f. 40 v b), and, later, a short biography of St Jerome, 
He mentions that he finished the MS of the Romans part on 6th 
May (f 80 vb). It is without the Canons of M and N, but appears 
to have all the other prefatory material. At Rom. v it gives the 
Cassiodorian insertions at the same places as C and F, but without 
the indication provided by these that the passages were not in the 
'exemplar.' I have refrained from collating this MS in full, as the 
specimens taken prove it to be secondary in character, and quite 
unlikely to contain anything of value not present in C or F. Yet 

1 Many of the alterations are changes of order of words like this, showing 
evidence of interest in style. 

2 See the Catalogue. 


a good scholar, with no other MS to aid him than this, could 
improve the published Pseudo- Jerome considerably \ 

Here ends our examination of the Pseudo- Jerome MSS. Though 
it has been long, it is yet in a sense merely provisional. The problem 
of the relation between the two recensions has hitherto baffled me. 
It is almost impossible to study it until all the interpolations are 
before one in print, and it may for ever defy complete solution. 

Cassiodorus (Pseudo-Primasius): Revision of Pelagius 

In the first chapter- attention was called to a very important 
passage in the Institutiones^ of Cassiodorus, where he describes a 
copy of a commentary on thirteen Epistles of St Paul in his library, 
which was in wide use and was falsely attributed to Pope Gelasius*. 
Finding Pelagian poison in it, he cleansed the Epistle to the 
Romans with what care he could, and left th« expositions of the 
other epistles in a 'chartaceus codex' (i.e. a papyrus book of the 
modern shape), for his monks to correct in the same way. The 
annotations thus attributed to Gelasius were undoubtedly the 
expositions of Pelagius. 

After he has enumerated complete commentaries on St Paul's 
Epistles he proceeds to mention commentaries on single epistles. 
It will be instructive to set down here, in the same pure text as 
before, the exact words he uses about commentaries on the Epistle 
to the Romans in his possession : 

§ 13. Sancti Pauli prima omnium et aramirabilior destinata cognoscitur 
ad Romanes, quam Origenes uiginti libris Greco sermone declarauit, quos 
tamen supra dictus Rufinns in decem libris redigens adhuc copiose transtulit 
in Latinum. 

§ 14. Sanctus uero Augustinus ipsam epistulam inchoauerat exponendam, 
in cuius tantum salutatione uuum librum se profudisse commemorat, et— ut 
eius uerbis utar — operis ipsius magnitudine ac labore deterritus, in alia faciliora 
deflexus est. 

1 The Dean of Wells turned it to profit in the paper mentioned on p. 41 n. 4. 

2 P. 15. 

* Written between 551 and 562 (P. Lehmann, in PItiloIoyus, Bd. lxxi [1912] 
p. 295). 

^ The names 'Gelasius' and 'Pelagius' were sometimes confused: e.g. in the 
MSS of the Decretum Gratiani (composed between 1139 and 1142) 'Gelasius' 
appears as 'Pelagius,' cf. Von Dobschiitz, Deer. Gelas. pp. 120, 192. 


§ 15. Qui etiam scribens ad Simplicianum episcopum Mediolanensem sublimes 
et ezquisitas de eadem epistula tractauit aliquas quaestiones, quas nos predicto 
codici iudicauimus inserendas, ne, dum ezpositio diuisa queritur, legentis intentio 
noxie dififeratur. 

A reference in his De Orthographia^ to works he had com- 
posed in monastic retirement is also important : 

post expositioiiem epistolae quae scribitur ad Romanos, unde Pelagianae 
haereseos prauitates amoui, quod etiam in reliquo commentario facere 
sequentes ammonui. 

The meaning of section 15, which no one prior to myself 
appears to have read with any care — else Cassiodonis's revision 
might have been identified about four centuries ago — , is that he 
had inserted in his depelagianised Romans some of the De Diversis 
Quaestionihus ad Simplicianum of Augustine. Clearly, then, if one is 
in search of Cassiodorus's revision, one must look for an exposition 
of the thirteen Epistles, which has extracts from Aug. De Div. 
Quaest. ad Simplicianum in the Romans commentary. Yet no one 
appears to have done so. 

Instead, we have Gamier suggesting that the printed Pseudo- 
Jerome is Cassiodorus's revision of Pelagius, and in spite of 
obvious difficulties this view held the field for over two centuries. 
The strong commonsense of Zimmer destroyed it, but his own 
view that the Pseudo-Primasius' is the Gelasio-Pelagian com- 
mentary as it reached the hands of Cassiodorus, is also impossible. 
Dr C. H. Turner, as we have said, first guessed the true situation, 
that Pseudo-Primasius is Cassiodorus's revision, and it was left 
for me to prove this view the right one, by producing the passage 
in section 15. There is only one commentary on the Pauline 
Epistles which has long extracts from the Div. Quaest. ad Simpl. 
incorporated in it, and that is the commentary published at Lyons 
in 1537 under the title ' Primasii | Vticensis in Afrijca Episcopi, 
in omnes | D. Pauli epistolas commentarij perbre]ues ac docti, 
ante annos mille | ab autore editi. | Nunc uero primum loannis 
Gagneij Theologi, ac j Doctoris regij opera in lucem emissi. j Apud 
Seb. Gryphium | Lugduni. | 1537. | Cum Priuilegio regio ad 

But, if this be so, how came Gagney to call it by the name of 
1 Praef. (Keil, Grammatici Latini, vol. vii p. 144). 


Primasius? This error can, I think, be explained. In the year 
1536, the year preceding the publication of ' Primasius,' there 
appeared at Strassburg the first edition of Zmaragdus's Expositio 
Libri Comitis. Amongst the authors Zmaragdus professes to have 
excerpted are Pelagius and Primasius. In his margins he generally 
employed P for Pelagius, PR and PRI for Primasius. Now the 
latter symbols never occur in Zmaragdus's MSS except in refer- 
ence to extracts from the genuine commentary of Primasius od 
the Apocalypse, while the former symbol occurs only in passages 
from the Pauline Epistles. It is clear, therefore, to us that 
Zmaragdus knew no Primasius on the Epistles ; but the first 
editor carelessly expanded P everywhere into 'Ex Primasio.' 
Gagney found a number of these extracts in the (anonymous) 
commentary he had discovered. He therefore not unnaturally 
assumed that 'Primasius' must be its author, and put his name 
on the title page. 

Gagney found his commentary, he tells us, in a manuscript of 
the 'coenobium diui Theuderici, apud oppidum...quod uulgo Sanc- 
tum caput appellant. Colonia est Viennensis archiepT, non procul 
Lugduno in Delphinatu.' This means * the religious house of 
St Thierry at St Chef in Dauphine, subject to the Abp of Vienne, 
and not far from Lyons.' There can be little doubt that the 
St Chef MS, if it still existed, would be in the Grenoble library. 
But it is not there, and for its text we must now depend on the 
editio princeps^. 

But there is an (anonymous) MS of the commentary at Grenoble, 
which was formerly in the Grande Chartreuse, and this, strange to 
say, appears to be the only surviving manuscript of it. The fact 
is strange, because there were several other copies in existence in 
the ninth century. The following commentators certainly pos- 
sessed copies, for they made use of the Cassiodorian compilation ; 
Zmaragdus of St Mihiel (between 819 and 830), under the symbol 
P; Claudius of Turin (between 815 and 820), under the symbol 
ANT- (apparently = ANON^) ; Sedulius Scottus of Liege and the 

1 Eeprinted at Cologne in 1538, at Paris in 1543: later reprints precede that in 
Migne (P. L. lxviii), which is much less accurate than the editio princeps. 
- In a MS of his comm. on Tit. Philem. Hebr. (Paris 10,878 [saec. ix]). 
^ Hatto of Vercelli's commentary on these epistles appears to be identical with 


Rhine country (between 848 and 859) as ISID; Haymo of 
Auxerrei (about 850?). 

The mention of this Haymo is of especial interest, because the 
commentary on Hebrews which forms part of the published 
Pseudo-Primasius, has nothing to do with Cassiodorus, and is in 
fact the production of Haymo, monk of Auxerre^ Cassiodorus dis- 
tinctly says that the Pelagian commentary he had in his hands 
concerned 13 epistles. The fact is that when a commentary on 
Hebrews was required to complete the set of Epistles, [Haymo ^] 
was attached to [Cassiodorus], in the same way that [Alcuin] was 
attached to [Ambrosiaster], and a commentary was added to 
[Pelagius] in the longer Pseudo-Jerome form. 

That Pseudo-Primasius is not really the work of Primasius of 
Hadrumetum, was suspected by some readers at an early date. 
In the Hort copy of Pseudo-Primasius, editio princeps, since his 
death the property of the Dean of Wells, we find two MS notes 
contemporary with the book to the following effect: 'Commentarios 
hos non esse Primasii duplici patet argumento. Turn quod trithe- 
mius huius operis non meminit Tum etiam quia Cassiodorum citat 
folio 526* Qui iuxta consentientem Historiographorum senten- 
tiam Primasio recentior est.' Further, the words {in Phil, i) sed 
etiam corda tetigisset qiiamuis in Actibus Legamus, fidem uohm- 
tariam esse: tanien are underlined, and we find in the margin: 
'Vide scolia titulo diui Hieronymi que uidentur hoc loco pelagium 
authorem habere non hieronymum, nisi hieronymum faciamus 
pelagianum.' We have seen also that Thomas Gataker, that 
miracle of learning, doubted Primasian authorship^ 

The St Chef and Grande Chartreuse copies both contain the 
Haymo on Hebrews, without any indication of difference of author. 

Claudius of Turin (see Eiggenbach, Die clltest. lateiii. Komm. pp. 2.5 — 33, as regards 

1 See Eiggenbach, op. cit. pp. 185 ff. ^ Eiggenbach, pp. 41 ft". 

^ I think I have seen all the old MSS of Haymo, and in none of these is there 
an author's name by the first hand. The copy of Haymo in Hebr. used to complete 
Cassiodorus was defective, having a lacuna at the end of c. iii. As the lacuna is 
indicated at the wrong place in the printed editions, instead of after Dominus ot fide 
(p. 709 1. 1), I was misled in my attempts to trace the MS of Haymo used. 

* On a passage of Hebrews. 

5 Adv. Misc. II 20 {Ojpera Critica, Utrecht IG98 fol.) p. 389 c. 
S. P. 21 


They therefore come from a common original not older than the 
middle of the ninth centuiy. Copies prior to that date, such as 
those in the hands of Zmaragdus, Claudius, Sedulius and Haymo 
himself, must have been without a commentary on Hebrews. 

The recovery of the original form of Pelagius and the identi- 
fication of ' Primasius ' as Cassiodorus have made it possible to 
study Cassiodorus's methods to perfection. The first thing to do 
is to underline in Cassiodorus all that has come direct from Pela- 
gius. It has long been recognised that Pelagius is abundantly 
represented in ' Primasius.' When once the Pelagian material has 
been set aside, one can study the remainder more easily. There is 
real Cassiodorus in the commentary on Romans, which has been 
carefully rewritten, but in the other epistles, the work of his 
pupils — perhaps not more than three in number — there is very 
little that is original. But most of the non-Pelagian part through- 
out is borrowed from other authors. The list of these authors and 
their works already identified, will add substantially to the cata- 
logue of the Cassiodorian library, compiled by A. Franz fifty j^ears 
ago\ The following identifications are my own, with two ex- 
ceptions : 

Augustine, De Diver sis Quaestiombus Lxxxiii. 

„ De Diversis Quaestionibus Ad Siynplicianum, pp. 

448—454, 477—4.87. 

„ De Genesi ad Litteram. 

„ Contra Academicos. 

Epistidae 140 ; 147. 

„ Contra Duas Epistulas Pelagianorum. 

„ De Natura et Gratia. 

„ De Peccatorum Meritis et Remissione. 

„ De Spiritu et Littera. 

„ De Praedestinatione Sanctorum. 

„ De Perfectione lustitiae Hominis. 

*Claudianus Mamertus, De Statu Animae. 
A Gallican (saec. v) commentator on the Psalms, p. 427 
(*Faustus Reiensis^). 

1 M. Aurelius Cassiodorus Senator, ein Beitrag zur Gesch. der theol. Lift. (Bresl. 
1872). I have asterisked authors not in Franz's list (pp. 80 — 87). 
- Suggestion of Dom Morin. 


Eucherius, Instructiones^. 

Jerome, Gomm. in Galatas. 

„ Apologia ad Pammachium (= epist. 48 [49]). 
„ Gomm. in Esaiam. 
*Prudentius, Apotheosis 918 — 919, p. 442, 11. 4 — 5-. 

Tyconius, Rules^. 

The value of Cassiodorus's revision for the restoration of the 
Pelagius text is very considerable, as we seem to possess it in a 
state of remarkable purity. In other epistles than Romans its 
text has real weight in deciding the text of Pelagius. In Romans 
Pelagius is of course handled with much more freedom, and is 
often rewritten in the interests of Augustinian teaching. Occasion- 
ally Cassiodorus modifies the Pelagian language rather than the 
Pelagian thought. 

In view of the well-known connexion of Cassiodorus with codex 
Amiatinus of the Vulgate, it is a matter of some interest to see 
what Cassiodorus has done with the Pelagian Biblical text^ This 
becomes perfectly clear, so far as Romans is concerned, by collating 
the Cassiodorian text with the Vulgate of Wordsworth and White. 
There is not the least doubt that Cassiodorus substituted the 
Vulgate for the Pelagian text. This Vulgate text was very close 
to Clm 4577 (saec. viii — ix^), a manuscript sometimes cited by 
W.-W. In fact I am tempted to think that this Munich MS con- 
tains a text extracted from a copy of the Cassiodorian commentary 
in the same way as MS 1163 (saec. Xii) of the Vienna library con- 
tains a text of Ephesians extracted from the genuine commentary 
of Jerome on that Epistle". Cassiodorus has, as might be ex- 
pected, done his work very carefully, but here and there he has 
by oversight allowed a reading of Pelagius (D type) to remain. 

1 See J.T.S. xiv (1912—1913) pp. 69—72; cf. also Porson's Letters to Travis 
(Lond. 1790) p. 351. 

- Identified by Prof. W. B. Anderson, University of Manchester, after I had in 
vain consulted other scholars. 

3 See J.T.S. xi (1909—1910) pp. 562 f. 

■* Cf. Dom Chapman in Revue Benedictine t. xxviii (1911) pp. 286 ff., who has 
made a considerable study of the subject. 

5 E.g. at Piom. i 31 absque honore is read by Clm 4577 and Cassiodorus, but by 
no other authority known to W.-W. 

^ De Bruyue's discovery, Revue Biblique, t. xii (1915) pp. 361 f. 



Nearly always the Grenoble MS gives a better biblical text than 
Gagney, where the two differ. Incidentally, this investigation is 
an excellent confirmation of the judgment of the Oxford editors 
of the Vulgate, who have made no use of Pseudo-Primasius in 
constituting their text. 

Something must be said with regard to the other Epistles 
also. The evidence is the same, so far as examined. I have taken 
at random 1 Cor. x, 2 Cor. iiii, Gal. v, Eph. iiii, Phil. ii. The 
differences from the Vulgate are few, and in almost all cases 
trifling. There can be no reasonable doubt, therefore, that Cassio- 
dorus's pupils, like Cassiodorus himself, intended to substitute the 
Vulgate text throughout. 

To pass to the comments, which are our main concern. What 
sort of text of Pelagius did Cassiodorus possess ? In the first 
place, it was absolutely without the Pseudo-Jerome iuterpolations. 
Did it then belong to the A family, or the B family, or to some 
otherwise unknown family ? The answer appears to be, rather to 
the B than to the A family. One or two examples will make this 

In Rom. i 30 ' elatus ' est qui effertur supra mensuras suas 
Cassiod., B : mensuram suam A. 

In Rom. ii 1 omnes qui huiusmodi iudicabant (indicant cod.) 
Cassiod., B (om. qui): omnes quidera iudicabant A. 

In Rom. ii 5 ad maiora abuteris uulnera Cassiod.: ad uulnera 
maiora uteris ABH. Here altered by Cassiod. 

In Rom. iii 1 reddit rationem quod lex inanis non fuerit, sed 
Cassiod. : reddit i-ationem quod lex contempta non prosit, et BH : 
reddita ratione quod lex contempta non prosit et A. 

In Rom. iiii 24 non ut sciremus tantum qualis fuerit, sed ut 
nobis in omnibus exemplo sit, ut filiis pater Cassiod.: non ut eius 
solum fidem sciremus, sed et eius quasi patris imiteraur exemplum 
H : non ut eius fidem solum sciremus (sciamus A), sed et ut eius 
quasi patris imitemur exemplum AB. 

Further instances of the character of the Cassiodorus text have 
already been given in other connexions^: and the printed text is 
good enough to render the use made of it by Zimmer, Riggenbach 
and Hellmann as profitable now as at the time of the publication 

1 Pp. 280 f., 243 etc. 


of their works. The agreement of Cassiodorus with any of our 
other witnesses gives us a sixth century reading which can never 
be safely neglected. If Cassiodorus stand alone, we may well be 
in presence of some deliberate alteration made in his scriptorium. 

The only fear that one sometimes has is that the authors of the 
Cassiodorian compilation occasionally consulted the Pseudo-Jerome 
in their hands. Yet it is more probable that they confined their 
attention to the Pelagius, and made no use of the Pseudo-Jerome 
they had. 

The Grenoble (Grande Chartreuse) MS 270 is of the end^ of 
the twelfth century. It is a very large MS, and the portion with 
which we are concerned consists of ten perfect quaternions (namely 
XX — XXVIIII), the folia of which are misnumbered, because 
four leaves have been overlooked. Fol. 76v is therefore really fol. 
80 V 2. All the preceding part of the MS has been lost, and is not 
at Grenoble. The MS never had any prefatory matter to Romans. 
It begins 'incipit epistola beati pauli ap(osto)li ad romanos ' 
with no name of author. The pages have each two columns. 
Cassiodorus in my view found no name on his Pelagius, and added 
no name to his revised Pelagius. 

A comparison of the editio prHncejJS^ with the Grande Chartreuse 
MS shows that the St Chef MS was a sister of the Grande Char- 
treuse MS. There can be no question of one of the two being a 
copy of the other*. Probably the Cassiodorian autograph, or an 
early copy of it, reached Lyons ^ where in the ninth century a 
copy of it was made, and the Haymo commentary on Hebrews was 

1 The catalogue 'saec. xii': Dr Holder, to whom I showed the MS, dated it 
more precisely as above. I have to thank the Paris authorities for enabling me to 
study the MS there, as also Monsieur Omont for permission to get the photograph 
of part of a page published in Proc. Brit. Acad. vol. vii, opposite p. 291. 

- Folia neglected between 7 and 8, 44 and 4-5, 60 and 61, 67 and 68. 

■' Dr H. A. Gibbons, Prof. W. B. Anderson, and above all my brother, John B. 
Souter, rendered me most valuable help in the collation of the editio princeps with 

■^ The proof of this and many other points connected with Cassiodorus is properly 
reserved for my Vienna edition of Cassiodorus. 

5 There is reason to believe that some of the Cassiodorian books came to Lyons. 
I believe Grenoble MS 197 (saec. xii) is descended from Cassiodorus's copy of 
Augustine Contra Duas Epistulas Pelagianorum (cf. J.T.S. xvi (1914 — 15) 
pp. 156 f.). 


added. The St Chef and Grande Chartreuse MSS are independent 
copies of this ninth century MS. What Gagney sent to press was 
a copy of the St Chef MS. he had caused to be made. An excel- 
lent edition of the Cassiodorian text can be made by the joint 
study of our two authorities, neither of which can be said absolutely 
to surpass the other. Hundreds of errors have crept into the text 
in the course of reprinting : the editio princeps is a good deal 
better than Migne. 


The MS which bears the signature Mp. th. f. 12 in the Univer- 
sity Library in Wiirzburg, is a copy of the Epistles of St Paul in 
Latin, written at the beginning of the eighth century in Irish 
minuscule. Its Irish glosses are among the most prized documents 
of the Celtic scholar^ But it also contains far more Latin glosses. 
These have been copied and studied b}^ Zimmer". As ' the glosses 
are, most of them, patently much later than the text^,' they are 
not likely to have as much value as they had twenty years ago, 
when the text of the original Pelagius was still undiscovered. Yet 
they are not without value. Zimmer has shown that they are taken 
from Pelagius in one form or another, Origen-Rufinus in Bom., 
Jerome in Gal. Eph. Tit. Philem., Augustine, Gregory, Isidore, 
and Hilarius (= Ambrosiaster) in. Rom. Most of them are labelled, 
but some are anonymous. By far the majority are called 'PI.' 
(i.e. Pelagius). An analysis of the notes on Rom. i 9 — 31 will give 
some idea of the character of the compilation. In all there are 
49 notes* on this passage. Of these 49, 27| are rightly labelled 
' PL,' as they are to be found in the uninterpolated Pelagius, but 
10| are wrongly labelled ' PL,' not being discoverable in the un- 
interpolated Pelagius. On the other hand there are 7^ anonymous 
notes, which are as a matter of fact to be found in the uninter- 

1 Cf. H. Zimmer, Glossae Hihernkae (Berol. 1881) pp. 1—198; W. Stokes aad 
J. Strachan, Thesaurus Palaeohibernicus, vol. i (Cambr. 1901) pp. xxiii ff., 499 — 712 ; 
R. Thiirneysen, Handbnch des Alt-Irischen, 2 Teil (Heidelberg, 1909) pp. 2—14 
(selections), etc. 

2 Pelagius in Irland, pp. 39 — 112, and passim; pp. 25 ff. above. 
^ Lindsaj', Notae Latinae, p. 493. 

•♦ I have not counted here five notes labelled 'PI,' which should have been called 
' HI' (Ambrosiaster), as Zimmer has shown (pp. 127 f.). 


polatecl Pelagius, while 3| anonymous notes are not found there. 
There are relics of a very good text of Pelagius in these glosses, 
e.g. in Bom. i 17 et ideo 'ex' et 'in' posuerit ut tautologiae uitium 
declinaret; i 21 a naturali sapore; ibid, recedens; i 24 cauteria et 
, combustiones. These are the correct readings, corrupted in many 
of our authorities. 

But in the first note, i 9 'PI.,' the words 'plena ueritate' come 
from Cassiodorus, and it is clear that we cannot rely on the 
presence or absence of a symbol as any indication of authorship'. 
The glossator also took the words or clauses that suited his purpose, 
and omitted those for which he had no use or no room. Further, 
he abridged even the matter he used, in the interests of space. 
It is therefore remarkable that there should be as much in the 
way of good readings as there is. 

It is perfectly clear that the compiler had both the original 
form "of Pelagius and the Cassiodorus revision in his possession, 
because there are places where both notes are given, though 
the one is ultimately derived from the other, for example : 

{In Rom. i 21 per insitam sibi rationem uel ex mundi factione 
(= Cassiod.). 
In Rom. i 20 per naturam uel per facture rationem (= Pelag.)". 

There is one matter of relationship to which reference must 
be made. There are real points of contact with Sedulius Scottus, 
where all our other authorities differ, for example : 

In 1 Cor. vi 20 Wb qui non est sui(?) non suam faciat uolun- 
tatem sed illius a quo emptus est. PI. sanguine Christi: sangui 
(sic) egrorum aliis nocet sanguis Christi mundum redemit. 

Ibid. Sedul. qui non est suus, non debet sibi uiuere, sed illi 
cuius sanguine emptus est.... non auro, non argento, sed san- 
guine Christi. sanguis aegrorum aliis nocet, sanguis uero Christi 
mundum redemit-*. 

In these two authorities, and apparently in these alone, these 
two notes are juxtaposed ; the thick type represents Pelagian 
material. The last part is un-Pelagian. The MSS of Sedulius 

^ For instance, in Rom. i 18, a real Pelagian extract is labelled 'Aug.' (Zimmer, 
p. 130). Similar instances on the same page and on p. 132. 

2 Cf. Zimmer, p. 124. 

3 The parallel is alluded to by Zimmer, p. 72 n. 


give GG (= Gregory) as a source in this neighbourhood, but Wb 
and Sd can hardly be independent of one another. I am quite 
ready to believe that the glossator used Sedulius Scottus, if the 
palaeographers will consent to date the glosses as late as his time\ 

In 1 Tim. iiii 1 Wb, spiritus qui doceat eum per se, id est, per 
Paulum semet ipsum, ut antiqui dicebant : haec dicit spiritus 
sanctus; post quam de misterio intimauit, indicat quod illud 
heriticorum nutibus {sic) obscuratur'-. 

Ihid. Sd, per ipsum scilicet Paulum prophetalem inducit 
affectum quo modo antiqui dicebant : haec dicit spiritus 

sanctus postquam de sacramento intimauit, nunc indicat quod 

illud sacramentum hereticorum nubibus obscuratur. 

It is hard to believe that Wb is not secondary to Sd here. 
Again the only Pelagian portions are given in thick type. Other 
instances of the same kind may be seen at Eph. iii 19^ and Philem. 
16^ etc. Hellmann, who recognises the relationship between W^b 
and Sd, would make W^b the uncle and Sd a nephews 

Whatever be the truth of their relationship, the value of Wb 
is merely that of an occasional makeweight, where there is some 
doubt as to the value of our main authorities. 

Zimmer also introduced us to the Vienna glosses (W^n). These 
are contained in MS 1247 (formerly Theol. CCLXXXVII, olim 49), 
which was written by Marianus Scottus, the founder of the Irish 
monastery in Ratisbon, between the middle of March and 17 May 
1079". This is also a MS of the Pauline Epistles. Here Pelagian 
glosses are introduced by 'P' or 'Pel,' or are anonymous. The 
total number of glosses is much smaller than in Wb, but there 
are in proportion more glosses of considerable length, and their 
text is purer and more accurate than those of Wb. I have chosen 
in Col. ii 6 — iii 12 as likely to afford the best test of the quality 
of the glosses. In this section there are 32 notes. Of these about 
9 are not really by Pelagius, though Q^ of these have 'P' attached. 

^ They appear to be of various dates. 

- The differences here from the text in Zimmer (p. 107) and Helhnann (pp. 165 f. ) 
are due to study of Stern's pliotograph, Epistolae Beati PauU... (Halle a. S., 1910). 
3 Hellmann, pp. 168 f. -> Hellmanu, p. 166. ^ Hellmann, p. 170. 

•^ Facsimile of a page in Chroust, ^loniunenta Palaeographica i Ser. x Lief. 1 Taf. 


Of the anonymous glosses 2 are by Pelagius, and 2 are not. 
In this MS also there are interesting points of contact with 

In Col. ii 11 Wn, 'P' quia peccata sepeliuntur in babtismo : 
tres unde babtismi tres dies in sepulchre significant. 

Ibid. Sd, quia peccata sepeliuntur in baptismo^ 

None of this is Pelagius. Augustine is the authority last named 
in the Sedulian margin. 

In Col. ii 15 Wn, sine timore...non latenter. 

Ibid. Sd, sine timore...non latenter. 
Neither of these two notes is Pelagian-. 

Zimmer estimates that only 21 ' Pelagius ' glosses are shared 
by Wb and Wn, while 15, anonymous in one, are called 'Pelagius' 
in the other^: so that Wn must be regarded as a witness inde- 
pendent of Wb. 

The number of manuscripts containing glosses ultimately or 
directly derived from Pelagius, must be very large. It is quite 
natural that not many have been discovered where Pelagius is 
used by name. Yet there are one or two in addition to Wb and 
Wn. Anonymous Pelagius glosses occur, for example, in Clm 9545 
(saec. x) (formerly of Altaich*). In Berlin Codex Phillippicus 1650 
(saec. X ex. — glosses Xl ex.), (formerly of St Vincent of Metz), a 
manuscript of the Pauline Epistles, glosses occur with the title 
'Pelg.' or 'Pelagius 'I But the vast majority of them are anony- 
mous, and Sedulius is also cited (by name). Even Lanfranc's name 
is attached to Pelagian material ; there seems no end to the com- 
plication in which these studies are involved. Clm 18530 (saec. 
XI — XII) (formerly of Tegernsee) is related to the Berlin MS, and 
enumerates at the beginning the commentators used : Lanfranc, 
Augustine, Ambrose, Origen, Pelagius, Sedatius (= Sedulius)*^. 
Hellmann surmises, on the basis of information communicated to 
him by E. Steinmeyer, that as the Berlin and later Munich MSS 

1 Zimmer, p. 147 u., who points out that the second part of Wn is represented 
in Irish in Wb. 

2 See also Eph. iii 19, Col. i 18, ii 19 in Hellmann, pp. 168 f. 

3 Zimmer, p. 155. * Hellmann, pp. 152, 186 ff. 

^ Eiggenbaeh, JJnheachtet gehlieh. Fragm. p. 22; Hellmann, p. 183. 
* Hellmann, p. xv. 


along with Berlin theol. fol. 481, Einsiedeln 16, Karlsruhe Aug. 
Lxxxiii, contain Old-High-German glosses, these manuscripts may 
be related in other respects also ; in other words, they may contain 
Pelagian material in their Latin glosses ^ I have not pursued 
the matter. 

Claudius of Turin 
This section must be more of the nature of an appeal than a 
statement of facts. Commentaries on all the Epistles of St Paul, 
with the exception of First and Second Thessalonians, and First 
and Second Timothy^, have come down to us under the name of 
Claudius, Bishop of Turin, a native of Spain. He produced Galatians 
about A.D. 815, Ephesians and Philippians about a.d. 816, Romans 
about A.D. 816 — 820, and First and Second Corinthians about 
A.D. 820^ but of these the commentaries on Galatians and Philemon 
alone are published^, with one or two prefaces to others. Yet there 
are a good many MSS, and some of them at least are of superlative 
quality. I believe the following list is more complete than any other 
published*, and I have made a personal examination of all those at 
Paris and Rome: 

Monte Cassino 48 (saec. xi in.«) (Rom. 1, 2 Thess. 1, 2 Tim. Tit. 

Philem. Hebr.). 
Orleans (Fleury) 88 (85) (saec. ix) (all). 

Paris, B.N. 2392 (saec. ix in.) (ff. 147) (Rom. 1 Cor. 2 Cor. [down 
to xi 23])^. 
„ 2393 (saec. xi) (Rom., Hebr.). 
„ „ 2394 (saec. X*) ([five quaternions lost at the beginning] 
Eph. Phil. Col. Tit. Philem. Hebr.). 
„ 2394 A (saec. x) (Gal. Eph. Phil. Col. Tit. Philem. Hebr. 

1 Thess. 2 Thess. 1 Tim. [stops abruptly at iiii 10], 

2 Tim.)". 

1 Hellmann, p. xv. 

^ The commeutaries on 1, 2 Thess., and 1, 2 Tim. in Claudius MSS appear to 
be unaltered Ambrosiaster. 

3 The dates in Monumenta Germaniae Historica: Epist. torn, iv {—Karolini aevi 
II) (Berol. 1895) pp. 596 ff. ; Manitius, Gesch. d. lat. Lit. des MA. i pp. 394 f. 

* Migne, P. L. civ pp. 841—912, 911—918, after Mai. 

' Cf. Manitius, p. 395. *> The date from Lowe, Beneventan Script, p. 342. 

'' Possibly author's autograph. * See note 1 on next page. 

9 Add this MS to those in Delisle, Cat. des MSS t. ii pp. 405 f. 


Paris, B.N. 2395 (saec. x^) (Hebr. [latter part], 1 Cor. 2 Cor.). 
„ 10878 (saec. ix) (Tit. Philem. Hebr.). 
„ ,, 12289 (Fleury) (saec. ix) (Rom., 1 Cor. [quaternion lost 
between ff. 96 and 97], 2 Cor.). 
„ 12290 (Fleury) (saec. ix) (certainly Eph. Phil. Col. Tit. 
Philem. Hebr.). 
Rome, Vat. 5775 (Bobbio, written at Tortona) (a.d. 862) (1 Cor. 
2 Cor.)^. 
„ „ Reg. 98 (Fleury) (saec. xii) (Rom. 1 Cor. 2 Cor.^ Eph, 
Phil. Col. Tit. Philem.). 
(Vat. lat. 9530 and 9546 are merely the copies which Mai made.) 
Certain facts which have come to my knowledge, must be here 
very briefly mentioned. 

(1) Romans. He uses Aug. and Orig.-Ruf. On Rom. viii he 
has the same part of Aug. as Cassiod. has (Migne, PL. Lxviii 
pp. 460 D, 461 a). On Rom. ii there is Pelagian material: on Rom. 
xii he is identical with Cassiod. (p. 496 c), and there is a parallel 
also on Rom, xvi 16 'non ficto — saluatorem' (p. 505 d). I copied 
the entire text of Claudius's exposition of Rom. v from what 
appeared to be the best MS, Paris B,N. 2392. Cassiodorus alter- 
nates with some other source or sources. The following extracts 
occur from him: Migne P.L. Lxviii p. 437 11. 35 — 45; p. 437 
1. 55—438 1. 30; p. 438 11. 32—35; 11. 47—56; p. 438 1. 57—439 
1. 15; p. 439 11. 28—31; 11. 52—55; p. 440 11. 46—47; 11. 50—54; 
p, 440 1. 57—441 1. 39; p. 441 11. 50—51; p. 442 11. 54—56. The 
Cassiodorian notes on this chapter fill 327 lines of Migne, and 
of these 327 lines 135, or about five twelfths, have been taken 
over by Claudius. This chapter was selected by me merely because 

1 The date of these MSS must be saec. x, because Elisha the Archdeacon of 
Auxerre mentioned in them as owner, is i-ecorded in the obituary of Auxerre (Paris 
MS B.N. lat, 5253 (lat. 894) [saec. x— xi, Lebeuf dates MS about a.d. 1007]); 'xni 
Kal. Ian. [ = 20 Dec] ob. Eliseus archidiaconus, qui dedit fratribus suis res pro- 
prietatis sue, in uilla quae dicitur Gratiacus' [ = Grisy] (Recueil des Historiens de 
la France: Obituaires t. iii (Paris, 1909) p. 246); cf. Lebeuf, Histoire d' Auxerre 
t. I (Paris, 1743) p, 748; Max Quantin, Dictionnaire Topographique du Diparte- 
ment de L'Yonne (Paris, 1862) p. 68. I am deeply indebted to Monsieur Leon Dorez 
for these facts and references. 

^ A photograph of one page in Ehrle-Liebaert, Specimina no. 31. 

^ Catalogue errs in giving 'Gal.' also. 


of its interest in connexion with the incriminating passage of 

(2) First Corinthians. The preface consists of the Ambrosiaster 
preface interwoven with the preface that we have found in the 
second class of Pseudo-Jerome MSS. Claudius, however, may have 
taken the latter from a biblical MS. 

(3) Second Corinthians. The preface consists of genuine Ambst. 
+ [Pelagius] prologue + genuine Ambst.\ thus combining the whole 
of both. First part of commentary is genuine Ambst., as is also the 

(4) Ephesians. He gives the Cassiodorian prologue, but adds 
a piece to it. There is evidence of use of the Cassiodorian com- 
mentary also (e.g. p. 62.5 11. 14—18 (which = Pelagius); 11. 81—35). 

(5) Philippians has exactly the Ambrosiaster prologue. 

(6) Colossians has the Ambrosiaster prologue followed by 

(7) Titus has the Cassiodorus prologue. Paris MS 10,878 is the 
only one known to me that has the authorities indicated in the 
margin. It specifies IH, AN', AG (= Jerome, Anon., Augustine, re- 
spectively). It has AN.' seventeen times, and every time the passage 
so marked is from the Cassiodorian commentary. 

(8) Philemon has the Cassiodorus prologue. 

(9) Hebrews seems to have the usual biblical prologue. 

It is fairly clear even from this slight examination that Claudius 
exploited the Cassiodorian commentary, but there is no evidence 
known to me that he had either Pelagius or Pseudo-Jerome in his 
hands. That he was hostile to Pelagius is quite clear from the 
introductory letter to the Ephesians commentary^ 

With the commentaries of Claudius are closely connected those 
b}^ Hatto of Vercelli, edited by C. Burontius del Signore from the 
autograph of Hatto (which still exists^) in two delightful folios, at 
Vercelli itself in 1768. Riggenbach has shown that Hatto took over 

1 Cf. J.T.S. IV (1902—03) p. 90. 

2 (Best edited in) (M.G.H.) Epistolae KaroUni Aevi t. ii (Berol. 1895) p. 598 
11. 22—23. 

3 Vercelli, Bibl. Capit. 40 (xxxix) (saec. x) a page photographed in Monumenta 
Palaeographica Sacra (Torino: Bocca, 1899) no. xvii; the only other MS is Karlsruhe, 
Aug. CL (saec. x — xi), as Dr Holder informed me. But 1, 2 Cor. also in Bamberg, 
B. II 20 (saec. xi). The commentaries are reprinted in Migne, P.L. cxxxiv. 




the commentaries of Claudius on Philemon and Hebrews practically 
unaltered ^ I can testify that the same is true with regard to that on 
Titus. It would appear that Hatto is more independent of Claudius 
in the longer epistles, but that he used him throughout. Of Hatto, 
as of Claudius, it may be said that 1 and 2 Thess. and 1 and 2 Tim. 
are Ambrosiaster. Hatto once'^, however, uses Pelagius by name: 

In 1 Cor. XV 31: Pelagius dicit quia 'per' non semper pro iura- 
mento accipitur: nam cum dicimus: 'per puerum misi' et similia, 
non iuramenti est.' This is a paraphrastic reference to the com- 
mentary, which we remember was known at Verona in the Middle 
Ages^ Hellmann^ cites three passages »where Cassiodorus is used 
by him, but he naturally had no occasion to ask whether these 
passages come direct from Cassiodorus or through the mediation 
of Claudius. 

Zmaragdus of St Mihiel 

Zmaragdus' in his own preface (of date between 819 and 830) 
declares that he used 'Pelagius' in his compilation. There is no 
evidence that he did so anywhere else than in the Pauline Epistles. 
It is of some importance, therefore, to enumerate the passages of 
the Epistles in the Lectionary for the understanding of which his 
exposition was compiled. They are these, with the column of Migne^ 
added, where the exposition of each section begins : 


1 Cor. 

i 1— 6 

vi 3—11 

] 9—2.3 

viii 12—17 



X 10—18 

xii 1 — 5 



xiii 8—10 


XV 4—13 

i 4— 8 

iiii 1 — 5 

V 7— 8 




















1 Cor. 

viiii 24-x 4 




X 6—13 


vi 10 


xii 2—11 



i 3— 8 


xiii 1 — 13 


iii 13—21 


XV 1—10 


iiii 1— 6 


2 Cor. 

i 3— 7 




iii 4— 9 


V 1— 9 


vi 1—10 




viiii 6—10 


vi 10—17 




i 6—11 


xii 9 


ii 5—11 



i 11—20 


iii 17—21 


iii 16—22 



i 9—11 


iiii 1 — 7 


iii 1— 4 




1 Thess 

iiii 1— 7 


V 16—24 



ii 11—15 


1 Die fdtesten lat. Kommentare pp. 25 ff. ^ Eiggenbach, p. 30 n. 

3 See p. 23. * P. 182, n. 7. ^ gee pp. 31 f. « P.L. en. 

^ For Hebrews passages see Eiggenbach, Die dltesten lat. Komin. p. 39. 


Zmaragdus's compilation is preserved in whole or in part in 
the following MSS\ Those marked with an asterisk I have myself 
collated for symbols, and in some cases also for the Pelagian 

Angers, 233 (saec. ix), second part only (from St Aubin). 

*Berlin, Lat. 695 theol. fol. 344 (saec. x) (from Werden). 
Boulogne-sur-mer, 2.5 (saec. x) (from St Omer) (see Pitra in 
Migne, P.Z. cii p. 1112 fif.). 

Cdrdoba, Mosque (Cathedral) Library 1 (olim 72), (saec. x) 
(about A.D. 960). 

*Einsiedeln, 39 (saec. ix) (from Reichenau?) (imperfect). 

*London, Brit. Mus., Additional MSS 21914 (saec. x) (imperfect). 

*Luxemburg, 135 (29) (saec. x) (from Orval). 
Madrid, Archivo Historico Nacional I (saec. Xll). 

*Munich, Clm 6210 (saec. ix) (from Freising)^ 

*Munich, Clm 6214 (saec. x) (from Freising). 

*Oxford, Bodleian, Barlow 4 (saec. ix — x). 

*Paris, B.N. 2341 (saec. ix) (from Reichenau?). 

*Paris, B.N. 12045 (saec. ix), defective at beginning and end 
(from St Maur-des-Fosses). 

*St Gall, Stiftsbibliothek 424 (saec. ix). 

*St Gall, Stiftsbibliothek 435 (saec. x). 

*St Omer, 257 (saec. x), second part only. 

*Zurich, Kantonsbibliothek xii (saec. x) (from Rheinau). 

*Zurich, Kantonsbibliothek xxxii (saec. x) imperfect (from 

For a detailed account of the symbols used throughout the 
work in the MSS, the reader is referred to my two articles^ Here 
it is enough to say that P was employed by Zmaragdus to indicate 
' Pelagius,' whether, as in the vast majority of cases, it be the original 

> This list is, I think, fuller than any other published : I owe two items to the 
kindness of Dom Wilmart. 

* The collation of this MS I owe to the courtesy of the ever lamented A. H. Kj'd, 
B.A., of Wadham Coll. Oxon. and the University of Manchester. 

3 Jonrn. Theol. Stud. vol. ix (1907—1908) pp. 584—597; vol. xxiii (1921— 1922) 
pp. 73 — 76. The Essai Critique stir la vie et les aurres de Siuaragde, thise soutenue 
par Louis Barbeau 2[) Janvier 1906 et jours suivaiits, has not, I believe, been published. 
I have to thank Dom Moriu and M. Henri Omont for information about it. I have 
not seen J. Schmidt in Der Katholik, Bd. lxxxvi (1906) pp. 241 — 257. 


form of Pelagius, or, as in a very few cases, the passages quoted are 
taken from the Cassiodorian revision of Pelagius \ As already 
mentioned^ the erroneous 'ex Primas.' of the editions everywhere 
represents a P (= ex Pelagio) of the MSS. It is not too much to 
say that Zmaragdus relied most of all on 'Pelagius' for notes on 
the Pauline Epistles, all the more perhaps because he was not in 
possession of Ambrosiaster^ In addition to Pelagius and Cassiodorus, 
he employed for the Epistles of St Paul, Origen-Rufinus on Romans, 
Jerome on Galatians and Ephesians, Chrysostom latinised, Cassio- 
dorus (apart from the revision of Pelagius), Isidore*, Cyril on 
Philippians latinised, Victor of Capua, Gregory, Augustine Be 
Diuersis Qiiaestionibus Lxxxiii, Pseudo- Augustine Quaestiones 
Veteris et Novi Testamenti cxxviiK 

Zmaragdus's method may be illustrated after Riggenbach^: the 
Pelagian portions preserved by Cassiodorus are in thick type. The 
text is naturally that of Zmaragdusl 

{In Rom. vi 19 = Zin. 406 c) P. Hoc est, maiora quidera exigere a nobis 
pro diuinae seruitutis retributione deberem, sed condescendens et temperans 
infirmltati uestrae humana et possibilia praedico atque suadeo, ut sicut prompt! 
fviistis ad sectanda noxia atque contraria, ita alacres sitis ad ea quae utilia et 
•saluti arnica sunt peragenda. 

Hoc] Id 2 codd. Zm. suadeo] persuadeo Cassiod. 

Clearly Cassiodorus only, and not Pelagius, was before 
Zmaragdus's eyes here. Pelagius differs very seriously from both. 

{In Rom. vi ll = Zm. 403 A — B) P. id est: sicut caput uestrum semel est 
mortuum, sic et uos membra illius eiFecti, uitae eius exempla sectamini, ut nihil 
morti ulterius debeatis, hoc est, ut iu uobis locum mors secunda non habeat. 
ille autem uiuit deo, qui Christi uestigia hurailitate, sanctificatione, pietate 

This is exactly Cassiodorus: of it Pelagius has hardly a word I 
It is clear, however, that in most places Zmaragdus used 

Pelagius rather than Cassiodorus. Let us take an instance where 

the two differ rather seriously: 

1 Riggenbacb, Unbeachtet gehliebene Fragmentc pp. 6 ff. ^ P. 320. 

3 The MS of Ambst. on Eomans now at St Mihiel (16) is dated 'saec. x,' and is 
therefore later than Zmaragdus's time. 

* In at least one case (p. 415) Isid. = Ps.-Ambr. De xlii Mans. 13. 
^ See my edition p. xxvi. ® Pp. 8f. 

' All three texts are, however, critically edited according to the MSS. 
8 Cf. also in Gal. v 18 (Zm. 449 d, Cassiod. 601a) (Hellmaun, p. 157). 


In Rom. viii 12 

(Zm. 411 B — c = Pelag.) Cassiod. (459 a) 

Hoc totum agit ut ostendat eis Hoc totura agit ut osteudat legem 

legem non esse necessariam, quae iieterem fidelibus non necessariam, 
carnalibus data est. quae carnalibus et peccatoribus data 


Hardly ever do we find any evidence of use of Pseudo- Jerome, 
but citations in the section in 2 Cor. vi 1 — 10 (Zm. pp. 118 f) are 
several times made from H, side by side with P. It is possible that 
the references to H refer to small portions of genuine Jerome 
sandwiched here and there between other borrowings: certain 
portions of this section have not yet been traced to their real 
author or authors. 

The MSS of Pelagius and Cassiodorus in Zmaragdus's possession 
were good. It is not often that Zmaragdus's theological pre- 
possessions lead him to contradict Pelagius, but they do on one 
occasion at least, in Rom. viii 29, where Pelagius says: 'praedes- 
tinare idem est quod praescire,' but Zmaragdus deliberately inserts 
the negative 'non' before 'idem.' Hellmann groups Zmaragdus as 
a textual authority with Sedulius and the Vienna glosses'. All are 
of good quality, but my own opinion is that Sedulius is, alike in 
extent and textual purity, the most important of the three. 

Sedulius Scottus 

Nothing is known of this scholarly Irishman before his appear- 
ance in Liege about a.d. 848. There he laid the foundations of an 
extended influence. After 850 he is found at Cologne, and he is 
known to have been in touch with Fulda, Metz and other places-. 
One of his strongest interests was the study of the Bible, on various 
parts of which he made collections. The 'Collectaneum in Epi- 
stolas Pauli' was edited by Johannes Sichardus as early as 1528 

1 Sedulius Scottus pp. 150 ff., 170. 

'^ Cf. Manitius, Geschichte pp. 315 ff.: on his works see also M. Esposito in 
Proceedings Royal Irish Acad. vol. xxviii (1910) section c pp. 62 — 95; Hermathena 
vol. XVI (1911) pp. 58—72, 329 ; Hellmann, Sedulius Scottus (Miinchen, 1906) ; Von 
Dobschiitz (Jahreslericht der Schles. Gesellschaft fiir Vaterl. Cultur 1913), 'Ein 
Biicherkleinod ' pp. 5 ff. 


at Basle, but the book is by no means rare\ His text was based on 
an old Fulda manuscript, and has since been reprinted several times'^. 
Sedulius or his school is now associated with certain Graeco-Latin 
Biblical MSS ; the Bibliotheque de I'Arsenal Psalter, the St Gall 
Gospels (A), and the Boernerianus (Gg) of the Epistles of St Paul. 

Though his influence was strong, his works have not been pre- 
served in many copies. Traube, and after him Hellmann^ could 
enumerate only five surviving MSS of the Pauline compilation ; 
that from which the editio princeps was printed has perished, like 
most of the great Fulda collection. These MSS are : — Clm 9545 
(saec. x) (from Oberaltaich), Clm 6238 (saec. x ex.) (from Freising), 
Zurich, Kantonsbibliothek, codex LXXii (saec. x) (from Rheinau), 
Fulda (saec. xi — xii) (= Weingarten 27), Bamberg B v 24 = Bible 
MSS 1 27 (saec. xi ex.*). In the opinion of Hellmann, the best MSS 
are those of Ztirich and Bamberg. I have therefore fully collated 
the Pelagian extracts in these, as also in the editio princeps. 
I have further collected the symbols indicating authorities used, 
in the two Munich MSS : the Fulda MS I have not seen. 

Even a careless reader of the first printed edition would see 
that Pelagius was one of the authorities employed by Sedulius, 
for near the beginning of the commentary there occur in the text 
(not in the margin) the words: 'Aliter secundum Pil(agium)^' 
But, as the first editor ignored the marginal symbols, which reveal 
that the work is only what it claims to be, a ' collectaneum,' even 
Zimmer declared that it quotes sources only here and there®. 
Hellmann's Sedidius Scottus revealed the real situation, and on 
the basis of a collation of the four MSS named I was allowed to 
publish 'The Sources of Sedulius Scottus' Coilectaneinn on the 
Epistles of St Paull' 

1 Best account of its genesis in P. Lehmann, lohannes Sichardus u.s.w. 
(Miinchen, 1911) pp. 54 f., 120. 

■^ Lastly Migne, P.L. cni. The orthography of the editio princeps is better than 
that of Migne, and a certain number of variations in reading have crept in since 1528. 

3 Pp. 190ff. * Hellmann, 'saec. xii.' 

^ The Irish spelling of Pelagius. The archetype had many Irish 'symptoms,' as 
a palaeographieal study of the MSS reveals. 

® 'Die (d. h. Pseudo-Primasius u. Sedulius) selbst iiberhaupt nur ganz vereinzelt 
Quellen citieren,' p. 112. 

' Journ. Theol. Stud. vol. xviii (1916—1917) pp. 181—228. 

S. P. 22 


This investigation proved extensive use of Pelagius, as the 
leading authority, under the symbols pelag, pela, pel, pilag, 
PILG, PIL. It proved also that he had used the following : Albinus 
(Alcuin) in Hebr.; Ambrosiaster on Rom. 1 Cor.; Theodore of 
Mopsuestia (in Latin) on Galatians; Augustine (various works) ; 
Pseudo-Augustine (Ambrosiaster) Quaestiones (the earlier edition); 
at least one Pseudo-Augustinian sermon; Basil; Bede; Cassian; 
Cassiodorus De Actihus ac Disciplinis, In Psaltenum; Eusebius- 
Rufinus Hist.-Eccl.', Faustus Reiensis; Gennadius; Gregory ; Jerome 
in Gal. Eph. Tit}, and a number of other works; John (i.e. Chry- 
sostom) ; Isidore, which means nearly always the Cassiodorus 
(Pseudo-Primasius) commentary (occasionally the Etymologiae)', 
Origen-Rufinus in Rom.; Junilius"^; Sedulius (about a column of 
original matter). 

To judge by the Pelagian extracts, Sedulius copied his sources 
most carefully. He has no knowledge of the Pseudo-Jerome, and 
so far as the Pelagian comments are concerned, he comes perhaps 
nearer to A (the Reichenau MS) than any other authority of 
which I have knowledge^ Unfortunately I did not make the 
same critical study of Sedulius's biblical text as I have devoted to 
the notes, but from a study of such hints of it as are obtainable 
from the printed text and my collations, I have the impression 
that Sedulius must have taken over for his lemmata the BD 
{Book of Armagh) type of text from his copy of Pelagius*. In 
other words, his procedure was to take Pelagius, text and com- 
mentary, as the basis of his Cullectaneum, omit from the notes 
what was unorthodox or useless for his purpose, and fill out its 
meagre, glossarial character from the contents of his library. I 
suggest therefore that a collation of the biblical text in the best 
MSS would be decidedly worth while, as a control over the B 
readings. It is quite clear that Sedulius did not use on this 
occasion the g text, as Traube at one time imagined. 

Study of Sedulius's Pelagian extracts has been rendered easy 

1 Never Ps.-Jerome on the Epistles of St Paul. 

* Jnnilius occurs in an uupublislied part of Sedulius I found in the Bheinau MS, 
but overlooked in writing the article just mentioned. 

^ Cf. 111 2 Cor. X 1 uindicandum ASdWb ; delicti GH: in Eph. vi 23 nam 
caritas AGSd; caritas H: in 1 Tim. iiii 8 aeuo ABSd; eo H, ; aeternum GH._>. 

* See W.-W.'s apparatus to Romans passim. 


by the most convenient fact that Hellmann restored their text 
critically from the MSS before publishing such as he used. He 
has had no difficulty in showing that it is a good text, much 
superior to Pseudo-Jerome and G, wherever it is available. 
Parallels between Wb and Sd have been already alluded to. They 
have un-Pelagian notes in common, e.g. at 1 Cor. vii 19 on circum- 
cisio nihil est and at 2 Tim. iiii 3 on prurientes auribus^. The 
contact with Wn is even more striking. Hellmann enumerates nine 
passages shared by Wn and Sd, not one of which is genuine 
Pelagius. One passage {in Col. i 2) shared by Wb and Wn, is also 
un-Pelagian. Facts like these suggest that some at least of the 
glosses in these MSS come from Sedulius. In Hellmann's scheme ^ 
Zm, Sd and Wn are represented as coming from one common root, 
and Wb is, so to speak, the uncle of all three. 

Of the Hebrews commentary in Sedulius the reader may be 
referred to Riggenbach's accounts 

Haymo of Auxerre 

Exegetical material, attributed in print sometimes to Haymo 
Bp of Halberstadt, sometimes to Remigius of Auxerre, has now 
been successfully vindicated for Haymo, a monk of Auxerre*. The 
oldest MSS, of the Pauline commentary at any rate, are anony- 
mous. It appears to have been composed about the middle of the 
ninth century, and was first printed in 1519 at Strassburg'. The 
following MSS have come to my knowledge, though I should say 
I have made no effort to record any later than the eleventh 
century : 
Paris, B.N. 2409 (saec. ix) (Rom. 1, 2 Cor. 1, 2 Thess.). 

„ „ 13409 (saec. ix) (1 Tim. [part only], 2 Tim. Phil. Eph. 
Hebr.) (first of St Peter's, Corbie, then of St Ger- 

„ „ 2412 (saec. x) (Hebr.) (under name of Ambrose). 

^ See Hellmaun, p. 159; but his other two examples are real Pelagius. 

2 P. 170. 3 Die altesten lat. Komm. pp. 212 ff. 

■* Riggenbach, op. cit. pp. 178 — 201. 

5 ' Haymo Sax. episc. Halberstatt. in diui Pauli epistolas cum breuis turn perlucida 
expositio : Excusum Argentinae per Renatum Beck Anno mdxix.' Copies appear to 
be very rare. 



Paris, B.N. 12303 (ancient no. 821) (saec. x) (2 Cor. 1, 2 Thess. 

1, 2 Tim. Phil, [also by a different scribe, Phil. 

1 Thess.] Eph.). This MS and 2412 are two parts 

of the same MS, once together in St Germain- 

„ 24.52 (saec. x). 
„ „ 17290 (saec. x) (Romans wanting; partly arranged as 

a lectionary). 
London, B.M. Harl. 3102 (saec. ix) (Rom. 1 Cor. down to xv 27 — 

28, rest lost) (a French MS). 
St Gall 333 (saec. x). 
Avranches 115 (saec. xi). 
Douai 343 (saec. xi): 344 (saec. xi) (Hebr.). 
Le Mans 229 (saec. xi) (Rom. 1, 2 Cor. Hebr.). 
Angers 67 (saec. xi): 1902 (saec. xi) (title and one leaf). 
Milan, Ambros. A 138 sup. (saec. ix — x) (Rom. Hebr. 1, 2 Cor.) 

Rome, Vat. lat, 615 + Vallicell. A. 8 (saec. xi — xii) (Rom. 1 Cor.)'^ 

(formerly of S. Euticio in Nursia ?). 
„ Bibl. Capitol. C. 102 (saec. xi) (lacks Tit. Philem.?). 
„ 103 (Gal. Eph. Phil. Col. Hebr.). 
Naples, Bibl. Nazionale, vi B 3 (saec. xii) (written in S. Lorenzo 

in Carminiano near Troja, between 1145andll65)^ 
„ Bibl. Nazionale, vi B 11 (saec. xi — xii) (formerly of Troja). 
Monte Cassino ff. 209 (pp. 420) 25 x 34 cm. (saec. xi in.). 

Having copied various portions of the British Museum MS, 
which is perhaps the very oldest of all the MSS, and compared it 
with the printed text, I can testify that Haymo is exceedingly well 
represented in print. But he is not of much use for constituting 
the text of Pelagius. He fished in most waters, and was an inde- 
pendent thinker besides. In Romans we can see use made of 
Ambrosiaster, Origen-Rufinus and Cassiodorus (Ps.-Primasius)^ 

^ See 'Dismembered Manuscripts' by the present writer in Revue Benedictine, 
t. XXIX (1912) pp. 367 f. 

^ See the article cited in the last note. 

* This superb MS like the other Naples and Monte Cassino MSS was kindly 
indicated to me by Dr Lowe (see now his Beneventan Script p. 322 etc.). 

* He has the tell-tale animositas at in Rom. i 29 (Pelag. intentio). 


I did not discover any evidence of direct use of Pelagius. But in 
any case he used his sources with considerable freedom, and much 
of his work seems to be original, at least in this sense that it is 
not derived fi-om other professed commentators on 'the Epistles. 
He appears on occasion to have used Zmaragdus (e.g. in 2 Cor. 
vi 5 is taken from Ps.-Aug. Quaest. V. et N. T. 120 through 


Isidore of Seville was the arch-compiler of the Middle Ages, 
and it can be shown that he extracted material from Pelagius 
among many other authors. 

Etym. vii 9, 9 '"electus," sicut in Actibus Apostolorum spiritus 
sanctus dicit (xiii 2): "Segregate mihi Barnaban et Paulum ad 
opus quod elegi eos,'" is from Pelag. in Rom. i 1 'sicut in Actibus 
Apostolorum spiritus sanctus dicit: "Segregate mihi Barnaban et 
Saulum ad opus quod elegi eos\" ' 

Etym.. X 46, ' " contentiosus " ab intentione uocatus, qui non 
ratione aliquid, sed sola pertinacia uindicat,' is from Pelag. in Rom. 
i 29 'contentio est, ubi non ratione aliquid, sed anirai pertinacia 

Etym. xi 2, 20 ' dicitur igitur " mulier " secundum femineum 
sexum, non secundum corruptionem integritatis : et hoc ex lingua 
sacrae scripturae. nam Eua statim facta de latere uiri sui, nondum 
contacta a uiro, mulier appellata est, etc' is from Pelag. in Gal. 
iiii 4 ' hie mulieris nomen non corruptionem, sed sexum significat, 
sicut et Eua statim ut facta est, mulier appellatur.' 

De Ecclesiasticis Officiis ii 18* 'et si non uis esse maior, esto 
uel minor, quia liberae uoluntatis es. nuptiae enim peccatum 
non sunt, sed per sollicitudinem mundi qui nubunt legem dei 

' Mr M. Esposito informs me that Isidore habitually uses the Vulgate (vg. here 
has adsumsi). 

^ Hellmann, p. 184, who gives other parallels also from the 10th book. 

3 This parallel I owe to Hellmann, but I have collated three MSS of Isidore ; 
Karlsruhe Augiensis ccliv (saec. viii — ix) f. 135, copied from a Visigothic original, 
St Gall 222 (saec. x) copied from an insular original, 240 (saec. ix): this last MS 
has lost some leaves (perhaps a whole quaternion) between pp. 280 — 281, words from 
uictitantes aut pane solo (Migne, P.L. lxxxiii 795 a) — in nouo autem testamento 
(807 c). Neither the first nor the second MS seems particularly good, nor can one 
be said to be really better than the other, to judge by this passage. 


seruare uix possunt. Aliter. Illos dicit non peccare si nubant, 
qui nondura uouerint deo castitatem. ceterum uel qui in corde 
suo promisit, si aliud fecerit, habet damnationem, quia primam 
fidem, sicut ait apostolus, irritam fecit, quod enim erat per 
naturam licitum, per uotum sibi fecit inlicitum, sicut Annanias et 
Saphyra, quibus de pretio possessionis suae retinere nihil licuit, 
ob quam causam et subita morte prostrati sunt.' This is from 
Pelag. in 1 Cor. vii 28: Isidore adds dei to legem, per naturam 
before licitum, and touches up the order of words once or twice, 
but otherwise the correspondence is exact. 

Quaest de Vet et Nov. Test. 32 §§ 37, 38 (ed. Arevalo, torn, v, 
p. 255): 'Die mihi: dilectio uel caritas in quot modis consistit^? 
In iiii. Hoc est, primum in dei timore uel dilectione; secundum, 
sicut nosmet ipsos sic deum amemus; tertium, proximos; quartum, 
etiam inimicos. deum ergo plus quam nos diligere debemus, proxi- 
mum sicut nos, inimicum ut proximum ; et nisi deum primum 
dilexerimus, nosmet ipsos minime diligere poterimus.' This is 
from Pelag. in Gal. v 14 ' dilectio uel caritas quattuor modis 
constat : hoc est, in dei dilectione, quae prima est ; secunda, si 
nosmet ipsos secundum deum amemus ; tertia proximos ; quarta, 
etiam inimicos. deum ergo plus quam nos diligere debemus ; 
proximum sicut nos; inimicum ut proximum... nisi deum primo 
dilexerimus, nos minime possumus non peccando diligere....' 

Like Isidore, Luculentius had borrowed from Pelagius without 
acknowledgement. Passages are in Rom. xii 6, 8^; there are 
perhaps others. 

Note on Prologues or Arguments 
It is notable that while the Primum quaeritur prologue is 
found in nearly all Vulgate MSS, a complete set of Pelagian pro- 
logues has not yet been produced from a biblical MS. Dom de 
Bruyne, the leading authority on Latin biblical prologues, has, 
however, discovered a few traces of them '. They are -these : 

arg. Phil. Paris, B.N. 9380 (saec. viii — ix)(formerly of Orleans), 15180 (saeo. 
xii) (formerly of Foigny) ; Ste-Genevi^ve 10 ; Puy ; Brussels 2. 
arg. iThess. Paris, B.N. 9380; 15180; Ste-Genevifeve 10; Puy; Basle BIG; 
Brussels 67 ; 68 ; Engelberg, 245. 
^ Arev. consistunt. ^ Migne, P.L. lxxii pp. 815 ff. 

^ Revue Binidictine t. xxiv (1907) p. 263, where add non before sit in the note. 




1 Tim. 


2 Tim. 






Paris, B.N. 9380; 15180; Puy; Basle BIG; Brussels 42. 
Paris, B.N. 9380; 15180; Ste-Genev. 10; 18; Puy; Basle BIC; 

Brussels 2 ; 68. 
Paris, B.N. 9380; Ste-Genev. 10; 18; Puy; Brussels 68. 
Paris, B.N. 9380, 15180 ; Puy ; Engelberg, 245. 
Paris, B.N. 9380 (partly). 

It is understood that Dom de Bruyne will edit all known 
Latin biblical prologues in the Collectanea Bihlica Latina. It has 
not been possible for me to collate any of these MSS except Paris 
15180; the great Theodulfian Bible was not accessible to readers 
during the war. In view of what has been written above, the fact 
that the oldest biblical testis for the prologues has Spanish con- 
nexions is interesting. 


{a) Textual Authorities 

The leading authorities are A and B, A especially for the extent 
of the commentary, B particularly for the biblical text. G, V and 
H come next in importance, so far as the commentary is concerned, 
for they contain practically the whole of it. The Cassiodorian 
revision is of almost equal value, subject to the qualifications that 
the language and the thought are modified, especially in Romans, 
in the interests of Augustinian theology, and a considerable portion 
of Pelagius is unused, while a Vulgate text is substituted in the 
lemmata. Sedulius Scottus used apparently the B type of Biblical 
text, and is a most exact quoter of the large portions of the com- 
mentaiy present in him. Other authorities, such as Zmaragdus, 
and the Wiirzburg and Vienna glosses, are on occasion useful as 
makeweights. In using all the post-Cassiodorian authorities, caution 
is required in distinguishing between the original Pelagius and the 
Cassiodorian revision, for both appear to have been accessible to 

(6) Citations of Pelagius hy Name 

The investigation in this and the second chapters has tended 
to show that the name 'Pelagius' might be attached to any form 
of the Pelagian commentary. Yet it is not without interest to 
tabulate here, chronologically and locally, such references to the 



[CH. VI 

Pelagian commentary hy name as have been discovered, mainly by 
other scholars. 





A.D. 412 and later 

North Africa 

Marius Mercator 

A.D. 429—431 

Constantinople, etc. 

Arnobius Junior 

about A.D. 450 


Anon. comm. in epp. canon. 


about A.D. 650 


Aug. ccxxxiii, saec. ix). 

Irish Canons 

about A.D. 700 


Book of Armagh 

A.D. 808 


Wiii-zburg, MS theol. F. 12 

saec. VIII in. 
(glosses by various 
scribes, of uncer- 
tain date) 



A.D. 809—830 

St Mihiel 

St Riquier catalogue 

A.D. 831 

St Riquier (Picardy) 

Murbach catalogue 

about A.D. 840 

Murbach (Alsace) 

Sedulius Scottus 

A.D. 848—858 

Liege and Rhine 

St Gall catalogue 

about A.D. 850 

St Gall 

Lorsch catalogue 

about A.D. 900 

Lorsch (s. of Darm- 

Hatto, Bp of Vercelli 

about A.D. 960 


Lucca plut. I 1 1 

saec. XI 


Vienna MS 1247 

A.D. 1079 


Berlin, cod. Phill. 1650 

about A.D. 1100 


Cod. Bodl. Laud. Misc. 350 

about A.D. 1100 

Eberbach (Bavaria) 

Clm 18530 

about A.D. 1100 

Tegernsee (Bavaria) 

Paris, B.N. 119291 

saec. XII 

Udine, Italy 

Cambridge, Corpus Christi 


48, E. 3 (saec. xii late) 

Eton MS 26 Bk. 3. 2 (saec. xii— | 


Cambridge, St John's Coll. 
(G. 15) (saec. xii — xiii) 

183 ^ 

about A.D. 1200 

St Albans, England 2* 

Dublin, Trin. Coll. A. 2. 2 (51 


bott) (saec. xii — xiii) (belonged 

saec. XV to West Dereham in 



Berne MS A 73 

saec. XIII 

France (?) 

John of Verona 

saec. XIII 


Milan, Bibl. Brera AE xiv 9 


saec. XIII 


1 De Bruyne, Eevue Biblique t. xii (1915) p. 383. 

* About this group I have learned from Dr James' Catalogues. I have collated 
the C.C.C. MS, and Professor John Fraser kindly copied the Dublin MS for me. 

PSEUDO- JEROME (pp. 268 f.; 276) 

The connexion between the Martyrologmm of Pseudo-Jerome and the 
Pauline commentary of Pseudo- Jerome (Hj, the shorter form) may be rather 
early and close. To the former ai'e prefixed two supposititious letters, one from 
Chromatins and Heliodorus to Jerome, and the other from Jerome to 
Chromatins and Heliodorus i. The reader will recall that the supposititious 
letter preceding the shorter form of the Pseudo-Jerome commentary is also 
addressed to Heliodorus. The view occurs to one that both these supposititious 
books, therefore, come from the same workshop. This idea is confirmed by 
the occurrence of the rather rare word lassescere in both the letters attributed 
to Jerome himself. If this hypothesis be correct, then the shorter form of 
Pseudo-Jerome is probably an Italian product, and it certainly antedates 
Cassiodorus, a view that we have seen to be probable from other considera- 
tions (p. 265) : for Cassiodorus actually refers to the letter of Jerome i)re- 
fixed to the Martyrologium^. 

1 Martyrologium Hieronymianum... ediderunt loh. Bapt. de Eossi et Ludov. 
Duchesne (Ex Act. SS., Novembris t. ii) (Bruxelles, n.d.) p. Ixxxii. 
^ Inst. c. 32, quoted by De Rossi and Duchesne, p. xlvii. 



abbreviations in MSS 205 f., 216, 221 ft., 
227, 230, 234 ff., 249 ft., 278 ff., 284 f., 
291 f., 296 ft., 306 ft., 309 f. 

Abelard 267 

ablative absolute, noun and present parti- 
ciple 81 

accents 300, 305 f. 

Acts of the Apostles, relationship of to 
the Epistles 71 

Ado, Abbot of Echternach 273 

Alcuin 302, 321, 338 

Ambrose 52 n. 5, 147 ff., 167, 172, 176, 
329, 339: for Pseudo-Ambrose see 
Ambrosiaster and Pseudo- 

AmbroBiaster 51 ff., 79, 85, 116, 120, 
134 ff., 152, 158, 167, 176 ff., 185, 
194 f., 261 f., 272, 321, 326, 330 n. 2, 
332, 333, 335, 338 his, 340 f. 

Ammian 200 

Amorbachs 274 ff . 

Ananias and Sapphira 70, 342 

Anglo-Saxon text 277 

Apollinaris 67 

apostles' supernatural gift 72 

arguments, authenticity of 115 : see 

Arians 67 

Armagh, Book of 17 n. 5, 25, 28, 126, 
132 bis, 134, 137 ff., 146 ff., 155 f., 222, 
245: see also under Dublin (Index 4) 

Arnobius 200 

Arnobius Junior 5, 266 

Athenagoras 267 n. 5 

Aubertiu 26 n. 4 

Augustine 4, 30, 36 ff., 117 n. 3, 119 f., 
185 ff., 194, 199, 239 f., 256, 261 f., 
318f.,322, 325n.5, 326, 329 feis,331f., 
335, 338 : for Pseudo-Augustine see 
authorities used by Pelagius 67, 174 ff. 

Barnabas and Saul 70, 341 
Bartholomew of Andlau 302 
Basil 338 
Batiffol, P. 187 n. 1 

Bayeux 286 n. 1 

Beauchamp, William 283 

Bee 286 n. 1 

Bede 338 

Beeson, C. H. 254 

Bellarmine 9 

Berger 138 f. 

Bernard of Clairvaux 311 ff. 

Biblical texts used by Pelagius 116 ff. 

Bobbio 216, 222 

Bornemann 24 

Bruyne, D. de 25, 156 f., 270, 342 ff. 

Buonaiuti, E. 178 n. 1 

Caelestius 266 

Caen, 286 n. 1 

Callimachus 267 

cancer 199 

canons: Irish 18, 28: Pelagian 269 f., 
301 f. 

capitula229f.,270, 277 f. 

Cassian 338 

Cassiodorus: (general) 14 ff., 22, 26 f., 
29 ff., 117, 265, 317 ff., 335, 338, 345: 
(Pseudo-Primasius) 32, 60, 63, 243, 
254, 258 n. 2, 263 f., 271, 280, 314 f., 
317, 318 ff., 327, 331 ff., 340 

Catharinus 7 

Catholic Epistles, anon, commentary on 
5, 344 

Cavallera 158 

Cave 17 

Ceillier 22 

Charlemagne 247, 302 

Chrysostom 193 ff., 335, 338 

Cicero 200 

Clairvaux 311 f. 

Clark, C. U. 254 

Claudianus Mamertus 322 

Claudius of Turin 51 n. 2, 272, 320, 322, 
330 ff. 

Clement of Alexandria 267 

Le Clerc 21 

Cologne 274 

Coutances 286 n. 1 

^ The more important references are indicated in thick type. 



Cyprian 128 n., 160 bis, 162 n. 1, 175 n. 3 

bis, 240 
Cyril of Alexandria 335 

Delisle, L. 330 n. 8 
Diodorus of Tarsus 196 

ecclesia defined 72 

editio princeps 281 f. : see also Amorbachs, 

Elisha: (Welsh king) 273: (archdeacon 

of Auxerre) 331 n. 1 
ellipsis 84 
Epimenides 267 

episcopus, one to each ciuitas 71 f. 
Erasmus 6, 268, 274 ff., 281 
Ettingen 276 
Eucherius 323 
Eusebius of Emesa 312 
Eusebius-Rufinus 338 
example, influence of on conduct 69 

Fabricius, J. A. 22 
Faustus Reiensis 322, 338 
Filaster 184 n. 4 
Fitzjames, Bp Richard 223 
Fontenelle 286 n. 1 
foreknowledge 70, 179, 192, 194 
Franz, A. 322 
freewill 188 
Freiburg i. Br. 274 ff. 
Freiburg fragments 229 ff. 
Fulgentius 73 n. 1, 256 

Gagney, Jean 319 f. 

Garetius 8 

Gamier 13 ff. , 24, 319 

Gataker, T. 26 n. 8, 321 

Gelasius, Pope 318 

genitive plural -unt 80 

Gennadius 338 

German history, document of 274 

gerund, ablative of 80 f. 

Gildas 142 ff. , 167, 169, 245 

Glossa Ordinaria 267, 275: see Walah- 

fridus Strabus 
glossary, Abolita 254 
Gospels' text in Pelagius 167 ff. 
Gothic 274 

Gray, Bp William 213, 215 
Gregory of Elvira 147 
Gregory of Nazianzus 240 
Gregory the Great 30, 239 f., 326, 328, 

335, 338 
Gregory, C. R. 24 
Gwynn, J. 123 n. 1, 139 f., 143 n. 1 

Haiswasser 294 

Hatto of Vercelli 320 n. 3, 332 f. 

Haussleiter, J. 26, 224 

Haymo of Auxerre 26 n. 8, 27 n. 7, 321 f., 
325, 339 ff. 

Haymo of Halberstadt 339 

Hebrews, Epistle to the, and commen- 
taries thereon 240, 242, 248, 268 f., 304, 
313, 315, 321, 325, 338 f. 

Hedio, Caspar 32 

Heinrich, German king 274 

Heliodorus 268, 281, 345 

Hellmann, S. 28 n. 2, 31 ff., 240 f., 263, 
324, 328 f., 333, 336 ff., 341, nn.2, 3 

Henry II of England 283 

heptateuch, Lyons 160 

heretics, 66 f. 

Hervaeus Burgidolensis 267 

Hieronymus : see Jerome 

Hilarius : see Ambrosiaster 

Hilary of Poitiers 147, 175 n. 3, 176 

Holder, A. 201 ff. 

Horace 200 

De Induratione Cordis Pharaonis 141 f., 
178 n. 1 

Instantius 147 

Ireland 236 

Irish MSS 237 etc. 

Isho'dad of Merv 195 d. 1 

Isidore of Seville 33, 184 n. 4, 199 n. 2, 
254, 326, 335, 338, 341 f. : for Pseudo- 
Isidore see Pseudo- 

Jansen 13 

Jerome 30, 63, 73 n. 1, 85 n. 1, 117, 120 f., 
138 u. 3, 146 n. 1, 157 f., 161, 168 n. 1, 
181, 183 ff., 214, 224 f., 239 f., 255 f., 
265, 267, 268 bis, 270, 275, 290, 293 f., 
317, 323, 326, 332, 335 f., 338: for 
Pseudo-Jerome see Pseudo- 
Jesuits : their first library at Paris 294 

John Chrysostom : see Chrysostom 

John the Deacon 61 ff. 

John of Verona 23 

Jovinian 4n. 6 

Jovinianists 67 

Julian of Aeclanum 140, 266 

Jumi^ges 286 n. 1 

Junilius 338 

justification by faith 70 

Juvenal 200 

Klasen 17, 24, 69 
Koetzschau 233 n. 2 

Labbe 10 n. 1, 14 n. 8 
Lactantius 175 n. 3 
Lagrange, M.-J. 157 f. 
laity 72 

Lanfranc 329 bis 
Latini, Latinio 132 
Lehmann, P. 254 n. 1 



Lightfoot, J. B. 24, 183 n. 4, 195 

Lindsay, W. M. 205 n. 1 etc., 254 

Lisieux 286 n. 1 

Livy 73 n. 1 

Loofs, F. 24 f., 69 

Lorsch 28, 302 

Lucifer of Cagliari 263 

Lucretius 199 f . 

Luculentius 342 

Lyons 325 

Macedonians 67 

Macrobius the Donatist 166 

Mangenot, E. 157, 184, 204 

Manicheans 67 

Marcion, Marcionites 50, 67 : Marcionite 

prologues 181, 223, 242, 269 
Marianus Scottus 328 
Martianay 21, 281, 282 nn. 
martyrologium ' Hieronymianum' 275 ff . , 

Matthew, apostle's career 71 
Maugerard, J.-B. 277 
medical illustrations 72 fi. 
Mercati, G. 48 ff., 226 2. 
Mercator, Marius 4, 41 ff. 
Merian 273, 280, 286 
Migne 282 

Mont St Michel 286 n. 1 
More, W. 317 
Morel, J.-B. 23 
Morin, G. 25, 31, 61, 141, 224 
Moyenmoutier 302 ff . 
Murbach 301 ff. , 305 

Nettleship, H. 208 n. 3 
Noris, Enrico de 13 
Novatian 175 n. 3 
Novatians 67 

Old-Latin Bible quotations, found only 

in Pelagius 161 n.2, cf. 166 
openings of notes, favourite 85 
order of words 83 

Origen 119, 174, 182 n. 1, 233, 267, 275 
Origen-Rufinus on Romans 5, 86 n. 2, 

188 ff.. 261 f., 318, 326, 329, 331, 335, 

338, 340 
orthography 206 ff., 209 ff., 214 ff., 236, 

238, 251 ff., 278, 283, 300, 306 

Pannenides 267 

participle present, genitive singular of 

Pauline Epistles, text of 119 ff. 

Pelagius : the name 1 f. ; Briton or Irish- 
man 2 f., his expositions of 13 Epistles 
of St Paul 3 ff. and passim ; on alle- 
gory 188; alternative explanations 
65 f. ; autograph copy 255 ; his biblical 

texts 116 ff.; Pauline text, latinity of 
153 ff. ; references to variae lectiones 
by 120 f. ; relation to Vulgate of Pauline 
Epistles 155 ff. ; cross references 64 f. ; 
date of commentary 4 n. 6; negative 
method of exegesis 66 ; attitude to the 
Epistle to the Hebrews 171 ; quoted 
by name 343 f . ; community of ideas 
throughout expositions 69 ff. ; prefaces 
and arguments 17, 117 f., 242, 247 f., 
269 [see also prologues); sections of 
Epistles 67 f. ; sources of expositions 
mS; Epistulaad Demetriademl5n.l, 
140 f., 158, 200 n. 2; De Libera Arbi- 
trio 256 

Philo 73 n. 1 

Photinians 67 

Du Pin 18 f. 

Pitra, J.-B. 31, 334 

Pliny the Younger 200 

Pontius Maximus 312 

Powys 273 

Praedestinatus : see Arnobius Junior 

predestination 70, 194 

Primasius 31 f., 320: for Pseudo-Prima- 
sius see Cassiodorus 

Priscillian : see Instantius 

probationes pennae 228, 305 

progress, moral 70 

prologues {see also Marcion, Pelagius) : 
115, 268 f., 270, 272, 342 f. 

Prudentius 323 

psalter, Galilean 161 

Pseudo-Ambrose 335 n. 4 ; Pseudo-Au- 
gustine 33.5, 338 bis, 341; Pseudo- 
Isidore 338; Pseudo- Jerome 6ff., 29, 
32 f., 35 ff., 50 f., 60, 174, 239 ff., 245, 
255, 257, 264, 265 ff., 325, 332, 336, 
339, 345 : see Martyrologium ; for 
Pseudo-Primasius see Cassiodorus 

punctuation 305 

quaternions 202, 213, 226, 232 f., 246, 
272, 287 ff., 295, 304, 312 f. 

Ramsbotham, A. 188 

Reisch, Gregory 275 ff. 

Remigius of Auxerre 26 n. 8, 339 

Resch, A. 168 n. 2 

rhythm 83 f. 

Eiggenbach, E. 31 f., 240, 324, 332 f., 

335, 339 
rotographs in possession of author 48 n. 2, 

229, 249 n. 6 
Rouen 286 n. 1 
Rufinus 188 

Sabatier 161 n. 2 
St Gall 28, 276 
St Riquier 22, 28 



St Wandrille 286 n. 1 

salvation, gratuitous 69 f. 

Sanday and Headlam 24 

Scherrer, G. 232 

Schoenemann, C. T. G. 23 

scholastic philosophy, fragment of 273 

scripture, favourite verses of 74 fJ. 

Sedulius Scottus 14, 18, 22, 26, 30, 32 f., 

63, 129, 141 n. 1, 148, 241, 245, 254, 

320,322, 327 £f., 336 ff. 
Seneca the Younger 73 n. 1, 200 
Severianus 199 
Simon Magus 70 
Simon, Eichard 19 ff., 174, 193 f. 
Sixtus of Siena 7 
Smaragdus: see Zmaragdus 
Smith, Alfred J. 175 ff. 
Solomon (Welsh king) 273 
Spain and Spanish influence 236, 254, 

263, 271 f., 292, 300, 308, 310: see 

Steinmeyer, E. 329 
Stern, L. C. 328 n.2 
stichometry, Pauline 242 f., 269 ff. 
style and language 79 ff. 
subnotation 50, 268 
subscriptions 273 
Swete, H. B. 174, 194 ff. 
Symbol ix, date of, in MSB 313 

teaching defined 71 

tenses, sequence of 82 

Tertullian 73 n. 1, 175 n. 3 his, 200, 

Theodore of Mopsuestia 195 ff., 338 
De Thou, J. 294 

Tillemont 21 
Traube 204, 337 
Treves 274 
Trithemiua 321 
Turner, C. H. 30 f., 319 
Tyconius 323 

Ussher, Abp 17 

Vallarsi 6, 22 f., 184 n. 1, 282 

Vatican fragments 48 ff., 134, 212, 226 ff. 

Verona 333 

Veronese script 249 ff. 

Victor of Capua 335 

Victorinus Afer 122 ff. 

Victorinus of Pettau 224 

Victorias of Keate 9, 281 

Vienna glosses : see Index 4 

Virgil 200 

Visigothic archetypes 235 ff., 245; spel- 
lings 237 f., 253 f. 

Voss, G. J. 10 ff. 

Vulgate 33, 116ff.,155ff., 211, 243f., 
260 f ., 268, 323 f. 

Walahfridus Strabus 18, 204 

Weihrich 132 n. 2 

Westcott, B. F. 26 n. 8 

words and phrases, characteristic 92 ff. 

Wiirzburg glosses : see Index 4 

Zahn, T. von, 193 f. 

Zimmer, H. 24 ff., 233, 236, 239 nn., 

240 f., 263, 319, 324, 326, 328 f., 337 
Zmaragdus 31 f., 63, 320 /n's, 322, 333ff., 

339, 341 


Gen. i26 75 

iii 4, 5 76 

xlviiii 10 159 

Exod. xxii 20 159 

Leu. vii9 (19) 159 

xxvil2 159 

Deut. XXX 6 159 

xxxii21 121, 160 

1 Eegn. xvi 7 160 

3 Kegu. iiii (v) 25 (9) 160 

Ps. xiiil 76 

xxxi 5 160 

Ixxiii 19 160 

Ixxxxiii 12 160 

cxviiii 7 160 

Prou. v22 161 

viiii8 161 

xi26 161 

xiiii6 161 

xvii6 129, 161 

xviii 17 162 

XX 13 .. 162 

XXV 8 162 

xxvi 12 162 

Eccl. vii 5 162 

Cant. vl6 162 

Iobi21 76 

viiii28 162 

xxiiii 23 163 

Sap. v2, 3 163 

Eccli. iiii 8 163 

v4 163 

v8, 9 163 

xxi 1 163 

XXXV 11 163 

Os. iiii 5, 6 164 

vi 3 164 n. 1 

viii 4 164 

Ion. iiii 11 164 

Zacb. ii8(12) 164 

Mai. ii7 164 

Esai. 122 164 f. 

vii 9 75,165 

xlii2 281 

11 165 

Iiii 4 76 

liii7 165 

lxvi2 165 

Hierem. iii 15 165 

iiii 3, 4 165 f. 

viiii23 166 

xvii 16 166 

Hierem. xxiii 24 166 

Ezech. xviii 20 166 

xxxiii 11 166 

Dan. ii47 166 

viiii23 166 

Math, v 39 75 

v44 75 

vii23 76 

X 10 76 

xiii 22 75 

xiii 43 76 

xiii 52 77 

xviiii 21 77 

xxiiii 13 75 

XXV 41 77 

Marc, vii 8 77 

xvi 17 167 

Luc. vii 47 168 

x7 77 

xi46 77 

XX 36 168 

XX 38 168 

xxi 34 168 

xxiii 34 77 

loh. i3 77 

iii 5 77 

vi56 77 

viiii 39 169 

xii 35 169 

xviiS 169 

xvii 24 169 

Act. ii3 169 f. 

iii 17 170 

iiii31f 170 

iiii32 77 

iiii34f 170 

v5 77 

v41 74, 170 

xiii 2 75, 170, 341 

xiii 11 77 

XV 10 77 

XX 26-28 170 f. 

XX 28 75 

xxvi 19 171 

Eom.i 1 341 

i 3 241, 263 

i 4 241 

18 281 

19 327 

19-32 326 f. 

1 10 260 

i 10-32 244 



Bom. i 15 148 

i 17 327 

il8 327 n. 1 

i 19 281 

i 20 327 

i 21 V.V.V//"/.V 241,"243,"'263, 327 bis 

i21f 142 

i24 292,327 

125 142,292 

i27 139,260 

i28-32 142 f. 

i 29 263, 340 n. 4, 341 

130 324 

131 323 n. 5 

132 260, 281 

ii 331 

111 122, 324 

li2f 61 

114 292, 309, 315 

115 260, 324 

liSf 143 

ii8 64 

11 9 282 

11 11 292 

lillf. 143 

11 12 282 

il 14 148 

iiUf 140 f. 

il 17 292 

1124 260 

11 26 282 

II 29 260 

III 1 324 

iil9 281 

11120 75 

11124 61 

11125 260 

ilil 1 241, 263 

1111 4 260 

111117 280 

111118 138, 139 

liii 18 f 260 

111121 260 

llii 24 324 

V 331 

v3f 62 

v4 148 

v6 285 

v9 309, 315 

V 12-15 41ff. 

V13-21 244 

V 14 62, 271, 309, 314, 315, 317 

V 14f 257, 269 

V 15 35 ff., 242 

V 15 f 257 

V 18 260 

V 20 122, 309, 315 6is, 317 

v20f. 271 

vl2 143 

vl3 122 

Eom. vl 3-14 61 

vl4-7 148 

vl 6 77, 309, 315 

vill 335 

vi 13 76 

vi 14 241 

vl 16 142,148,260 

vl 19 335 

vii7 260 

vli8 49, 241 

vii9-15 48 ff., 134, 228 

vii 12 77, 122 

vil 13 260, 309 his, 315 Ms 

vii 18 266 

vii 22 266 

viil 3 122, 309, 315, 316 

viii 3-7 228 

viil 3-8 48 ff., 134 

viii 6 241 

viii 12 336 

viii 17 212 

viii 19 260 

viii 22 260 

viii 24 77, 260 

viii 29 336 

viii 30 142 

viil 32 142 

viii 38 f. 148 

viiii 1 78 

viiil 2 212 

viiii 3 143 

viiii 6 122 

viiii 10 243 

viiii 11 142, 261 

viiii 12 266 

viiii 16 35, 37, 39 f. 

viiii 18 142 

viiii 20 142, 268 

viiii 21 261, 269 

viiii 26 309, 315 

viiii 32 261 

viiil 33 309, 315 f. 

x4 261 

X 12 122 

X 15 261 

xl9f 261 

xi 1 309, 316 

xllf 149 

xi3 143 

xi4 316 

xi 11 127 

xl 12 261 

xil3 269 

xl 17 127, 265 f. 

xi23 122, 261 

xl24 265 f. 

xi25f 149, 261 

xi32 127 

xi34 261 

xii 331 



Kom. xii 2 127 

xii5 261 

xii6 342 

xii 8 342 

xii 9 139 

xii 13 120f. 

xii 16 282 

xii 17 127 

xii 20 261 

xiiil 241,261 

xiii4 149 

xiii5 261 

xiiiSf 149 

xiiiQ 127,261 

xiiilO 78, 122 

xiiil2 128, 143, 261 

xiiil4 143 

xiiii4 241 

xiiii 6 261 

xiiii22 241 

XV 4 128 

xv6 261 

XV 12 261 

XV 16 128 

XV 19 261 

XV 21 241, 261, 263 

XV 24 261, 282 

XV 25 317 

XV 30 138 

xvi 5 261 

xvilO 261 

xvil6 331 

xvi 17 122,261 

xvi 18 122 

xvi 24 241, 263 

1 Cor. i2 317 

i4f 149 

ilO 128 

i23f 122 

ii3 128 

iil6 123 

iii3 123 

iiilOff 143 

iii 18 143 

iii 18f 149 

iiii 11-13 149 

iiii20 258 

vl 76 

v4 123 

v7 143 

v8 123 

vlOf 143 f. 

vill 293 

vi 15 309, 316 

vi 18 281 

vil9 78 

vi20 327 

vii 7 123 

viil4 36f., 39, 41 

vii 19 339 

1 Cor. vii 26 282 

vii 28 342 

vii29ff ... 149 

vii 35 138, 139, 140 

vii 36 63 

vii 40 123 

viiilO 74 

viiii 5 128 

viiii 15-20 230f. 

viiii 24 123 

x4 281 n. 1 

xl3 128 

x22 121, 223 

X24-31 230f. 

x25 230 

x32 123 

x33 78, 123 

xi 271 

xi 2f 257 

xi4 139 

xi 10 128 

xi24 138 

xi32 123 

xi34 128 

xiill 76 

xii 23 149 

xii 28 78 

xii 31 2G8 

xiii2 123 

xiii 2-4 257 

xiii 5 78 

xiii 7 76 

xiiii 1 241, 263 

xiiii 7 200 n. 1 

xiiiiU 123 

xiiii 16 123 

xiiii 25 78, 123 

XV 11 128 

xv22ff 149 

xv24f 257 

XV 31 123, 268, 333 

XV 44-58 52, 134 ff. 

XV 49 123 

XV 51 268 

xv51f 149 

xvi 1 129 

xvi 1-24 62, 134ff. 

2 Cor. i-iiii 280 

il-6 52, 134ff. 

i3ff 149 

14 ; 129 

16 285 

1 13 285 

124 129 

iiiS 242 

1116 188 

1111 If 144 

vl5 123 

V 15 f 269 

vl6 292f. 



2 Cor. vi 1 129 

vi 1-10 336 

vi5 341 

vi9 129b/A- 

villff 150 

vi 14 150 

vi 16 150 

vii 4 129 

vii 11 123, 139 

viiii 13 129 

X 280 

X 1 241, 338 n. 3 

x4 241 

X 16 243 

xi 1 123 

xi2 123 

xi9, 10 21 n. 1 

xi 13ff 144 

xi23 123 

xii2 75 

xii3ff 150 

xiil2 257 

xiii 3 123 

Gal.i 12 78 

ii2 138 

ii4 124 

ii5 130 

iilO 212, 258 

ii 12 212 

iii 16 150 

iii27 124 

iiii4 341 

iiii 6 281 

iiii 7 : 78 

iiii 21 ff 150 

iiii 24 188, 293 

iiii 25 212 f. 

V 11 124 

V 14 342 

V 18 242, 335 n. 8 

v21 124 

v22 64, 150 

vi 1 124 

vi2 124 

vi 7 124 

vilO 124 

vil2 124 

Eph.il7 213 

il7f. 264 

i 18 213 

ii5 124, 138, 139 6fs 

ii 14ff 150 

ii 17 130 

ii22 130 

iii 8 25Sbis 

iii 18 241 

iii 18f 263 

iii 19 328, 329 n. 2 

iiii 7 213, 264 

iiii 13 223 

S. P. 

Eph. iiiil8f 144 

iiii 22 78, 213, 264 

iiii 24 130 

v 5 258 

v 7 258 

v8 124 

v 14 243 

v 17 f 144 

v22 78 

v27 78, 124 

v 30 76 

vi 6 124 

vi 12 151 

vi 15 258 

vi 18 139, 140 

vi23 338 n. 3 

Phil, i 1 78, 124 

i 8 130, 144 

i 20 242 

i 22 242 

i 23 : 124 

i23f 151 

i27 139 

ii3 75, 124 

ii 6 195 

ii6ff 151 

ii7 74, 264 

ii 7f 124f. 

ii8 76 

ii22f 242 

iii7f 151 

iii 12 125 

iii 13 130 

iii 17 125, 126 

iii21 125 

iiii 3 257 

iiii 15 131 

iiii 17 131 

Col. i2 339 

i8 132 

i 12 132 

i 13 ff 151 

i 18 329 n. 2 

i24 78 

iio 78 f., 125, 132, 134 

ii6-23 328 f. 

ii 11 329 

iil5 329 

ii 18f lolf. 

ii 19 329 n. 2 

ii21ff 152 

iii 1-12 328f. 

iii 3 125 

iii 5 133 

iii5f 145 

iii8 139 

iii9ff 152 

iii 10 263 

iii 11 126 

iii 15 121 




Col. iii 19 50, 65 

iii21 133 

iiiil7 126 

1 Thess. 16 131 

19 131 

115f. 125 

li5ff 144 

1114 70f., 125 

1112 131 

1115 257, 282 

illl2ff 145 

ulie 145, 241 

uulOf. 258 

uiill 131 

111115 310 

v23 131f.,316 

2 Thess. i 2 125 

16 125 

113 121 

il3f 152 

1111 125 

1112 125 

nil 132 

1116 lS2bi$ 

1119 132 

Hi 10 132 

1 Tim. 1 9 74, 126 

115 145 

114 141 

116 139 

II 10 133 

III Iff 145 

lil2 280f. 

iii 10 145 

111 15 133 

nil 1 328 

llii 8 21 n. 1, 338 n. 3 

11119 145 

vl9 133 

v23 79 

vl2 243 

vl3ff 145, 152 

vi9 133 

vil7ff 145 

vl 18 133 

2 Tim. 1 10 133 

1 15 264 

2 Tim. 117 133 

11 4 133f. 

ii4f 145 

ii 11 75 

1112 126 

1124 21 n. 1, 281, 293 

11 26 213 

HI Iff 146 

ui8 146 

11113 339 

11115 139 

11117 79 

111115 241 

Tit. 15 79 

1 12 267 

I 15 282 

116 126 

il8 146 

lil3fE 152 

iii 8 126 

Phllem. r2ff 152 

16 328 

17 282 

25 243 

Hebr. 13 79, 171 

vi4 171 

x34 70f., 79 

x36 171 

xil3 171 

xil25 171 

lac. iii 14 171f. 

1 Petr. i 19 172 

II 1, 2 172 

117 172 

1123 172 

1119 172 

2 Petr. ill 9 172 

lloh. ii 6 172f. 

1117 173 

1121 178 

Hi 2 75 

1H6 79 

HI 16 173 

HH 18 75 

iiu20 173 

Apoc. vi 8 79, 173, 178 n. 2 


a, ab 93 

accessio 93 

accommodo 253 

adhuc 208 

adiutor 131 

adiutorium 93 

adsumo 93 

adtendo 89 f. : see attendo 

aemulatio 93 

aemulor 93 

aemulus 93 

affirmo 253 

ago 115 

alibi 91 

aliquanti 93 

aliquis 93 

aliter 66 

alius 93, 175 f. 

alterutrum 94 

anathema 94 

ancilla 252 

annuntio 253 

ante (adv.) 94 

antea 94 

appello 106 n. 1, 253 

applaudo 253 

arefacio 94 

arguo 94 

assensus 253 

attempto 253 

attendo 253 

auctoritas 94 

audenter 94 

auditor 131 n. 3 

auxilium 93 n. 2 

baiulo 94 
baptisma 95 
baptismum 95 
blasphemia 95 
breuiter 95 

caducus 95 

caelestis 113 

caerimonia 95 

calumnia 95 

calumnior 95 

carissimus 153 

caueo 90 

causa 95 f., 95 n. 3 : see reddo 

cautela 96 

censeo 96 

cetera 92 

clarifico 154 
coheres 253 n. 2 
cohortor 96 
commemoro 96 
commoneo 96 
compello 115 
concludo 97 
concupiscentia 166 n. 2 
confirmo 115 bis 
conparatio 97 
conprehendo 97 
conprobo 97 
considero 90 
consisto 97 
consuetudo 97 
contemno 97 
contingo 97 
contra 86 
contrarietas 98 
contrarius 98 
conuerto(r) 98, 177 n. 2 
copulo 115 
corpus 115 
corrigo 98 
credo 98 
crimen 153 
cum (conj.) 253 

Danihelus 207 

definio 115 

denoto 98 

deprecor 115 

deputo 98 

deseruio 98, 122 n. 2, 133 

deuterosis 183 

diaconissa 98 

dialecticus 99 

digamus 99 

dilectus 153 

dimico 99 

distraho 99 

diuersitas 99 

diuersus 99, 175 f. 

doctor 99 

dominor 115 

donee 84 f., 184 

dono 99 

duco 115 

duplex 99 

dupliciter 99 

ecclesia 72 
efficio 99 



egeo 103 n. 1 
emendo 99 f. 
enim 154 
euidenter 100 
ex 80 

exemplum 109 
exhibeo 100 
existo 100 
exoro 100 
expeto 115 
expono 87 

fades 153 
fiducialiter 100 
figura 102 
finio 100 

firmamentum 133 
firmiter 100 
firmo 100 
firmus 100 
fons 115 
forma 100 
frequenter 111 
fundamentum 133 

genero 100 
glorifico 154 
Graecus 253 
gratiae 101 
grauo 101 
gula 115, 253 f. 

habeo 101 

liabitaculum 130 

haruspex 94 

hera 297 

hie {2}ron.) 88 f., 101. 104 n. 1, 153 

hie (adv.) 89 

hinc 88 

honorifieo 154 

honoro 154 

hospes 253 

idcirco 102 
ideo 102 

idololatria 206 f. 
idoneus 103 n. 2 
lesus 208 
ille 104 n. 1, 153 
imago 102 
impedio 102 
impugno 102 
incautus 102 
incipio 87 

incorruptibilitas 102 
incredulus 103 n. 2 
increpo 115 
indebite 102 f. 
indebitus 102 f. 
indigeo 103 

induratio 141 
inebrio 103 
inferius 92 n. 1 
infero 103 n. 3 
infidelis 103 n. 2 
ingratus 103 his 
inhaereo 115 
inpossibilitas 103 
inremediabiliter 103 
inreprehensibilis 153 
inrogo 103 f. 
insensibihs 104 
ipse 153 
is 104, 153 
Istrahel 207 
ita : see ut 
iterum 115 
iubeo 104 
iuxta 112 

laesio 104 
lassesco 345 
legalis 104 
liber tas 104, 115 
ligo (verb) 104 
littera 104 
locus 89, 104 

maculo 105 

magnifico 129, 132, 154 

massa 178 

melior 80 

mens 115, 154 

mensura 105 

merces 105 

mereor 105 

minime 253 

modo 115 

monstro 86 n. 4 

monstruosus 105 

moralis 105 

mulier 181 

multus 175 f. 

mundus 153 

raunio 105 

nam 154 
ne 86 f., 90 
nee 105 
necesse 105 
nequis 86 f. 
nitor (verb) 105 
nobilitas 105 f. 
nomino 106 
noto 85, 90 
noui 106 
nullus 106 
nuncupo 106 

obiectio 106 





obliuio 106 
obsecro 253 
occasio 106 f . 
opto 107 
ostendo 86 

paganus 107 
parco 107 
participor 231 
pasco 107 
passibilis 107 
per 85 
perfecte 107 
perfectio 107 
perfectus 107 
perficio 107 
permaneo 115 
l^ersona 107 f., 153 
plus 115 
portendo 108 
possum 90 
postmodo 108 
postmodum 108 
praecipio 115 
praeiudico 108 
praemium 105 
praepostero (verb) 108 
I^raeposterus 108 
praesens 108 
praeualeo 108 
praeuenio 115 
pressura 154 
principor 108 
prior 108 

profectus (tiouii) 108 f. 
proficio 109 
propheta 92 
propono 109 
proprie 109 
prouoco 109 
pulchre 109 
puto 87, 109 f.,J.15 

quaero 86 

qualis 110 

qualitas 110 

quantus 110 

quautuslibet 110 

quantusuis 110 

quasi 110, 154 

quemadmodum 154 

querella 153 

qui : see quo 

quia 82 f., 154 f. 

quidam 175 f., 181, 183, 186 

quis (dir. interrog.) 90 

quisqae 110, 114 n. 3 

quo 110 

quoad usque 111 

quod 82 

quo modo 154 : see ut 
quoniam 154 f. 

rationabilis 111 
rationabiliter 111 
recapitulo 85, 127 
reddo 89, 111 
Regna 84 
reliquus 92 
I'ependo 111 
repeto 87 
replico 111 
respondeo 90 
ritus 111 

saepe 111 
sane 111 
satio (verb) 111 
scilicet 111 
scriptura 92 
sector (verb) 111 f. 
secundum 112 
sensus 112, 154 
sermo 155 
si 85 
sicut 154 
similo 112 
simul 90 f. 
siue 65 f . 

solacium 180 n. 2 
solum 112 
solum modo 112 
subaudio 112 
subdole 112 
subdolus 112 
subintellego 112 
suffragium 93 n. 2 
suffragor 115 
superflue 112 
superfiuo {adv.) 112 
superfluus 112 
super ius 91 f. 
su^jporto 253 
suscribo 112 n. 2, 252 
suscriptio 112 

tabernaculum 130 n. 1 
talis 89 
taliter 112 
tam — quam 113 
tamquam 154 
tango 113 
tantum 113 
tantum modo 112 
taxatio 113 
taxo 113 
tempus 113 
terrenus 113 
terrestris 113 
testamentum 84 



tolerantia 113 
tolero 113 
tollo 113 
totus 85 
tracto 114 
transitorie 114 
tribulatio 154 
triplex 114 
tunc 114 
typus 114 

uaco 253 
ualeo 114 
uelut 154 
uerbum 155 

uere 114 
ueritas 102 
uetus 80 
uices 111 
uidelieet 111 
uideo : see ne 
uindico 114 bis 
uindicta 114 n. 1 
uoco 106 

uolo : see ostendo 
usque dum 114 
usque quo 114 
ut 82, 114, 154 
uterque 114 
utor 115 


Angers 67 340 

233 334 

1902 340 

Avranches 115 340 

Bamberg A i 5 262 

B II 20 (Bibl. 89) 332 n. 3 

B v24 (Bibl. 127) 337 

HJiv 15 (Patr. 61) 15 

BasleB i 6 272 

Bi c 342f. 

Gil5 274f. 

Gii29 275 


lat. 695 (theol. fol. 344) 334 

theol. fol. 481 33, 330 

Phill. 1650 32, 329 

1831 254 

Berne A 73 28 

344 265 n. 2 

Boulogne-sur-mer 25 32,334 

Brussels 2 342f. 

42 343 

67 342 

68 342f. 

Cambridge, Univ. Library 

Ff iv31 271, 317f. 

Cambridge, C. C. C. Library 

48E 3 344 

Cambridge, St John's Coll. Library 

183 (G. 15) 344 

La Cava de' Tirreni, bibl. della 

Badial4 262 

Chartres 31 296 n. 1 

Cheltenham, Phillipps 518 51 

Colmar 38 303 

Cologne XXXIV 51 

Lviii (Darmst. 2052) 239 

ccxii 206 n. 2 

Cordoba, Mosque (Cathedral) Library 

1 (olim72) 334 

(Demidov (Russia) ?) 262 

Donaueschingen 191 207 

Douai 343 340 

344 340 

Dresden A 145" 262 f., 337 

Dublin, Trinity College Library 
Book of Armagh (Abbott 52) (see 
especially Index 1) ... 262, 323, 338 

A. 2. 2 344 

A. 4. 20 285 

254 265 n. 2 

Einsiedeln 16 33, 330 

39 334 

131 296 n. 1 

Engelberg 245 342 f. 

Epinal 6 271 f., 301f., 303ff. 

45(?) 272 

68 300, 302 

78 302 

Eton 26 Bk 3. 2 344 

Florence, Mediceo-Laurenziana 

Am. 1 J 262 f., 323 

Plut. xvDext. Cod. i 316 f. 

Plut. xviii Dext. Cod. ix 239 

Freiburg, Stadtarchiv 229 ff . 

Fukla, Bonifatianus 1 116, 262 f. 

Weingarten 27 337 

Gotha, membr. i 20 302 

membr. i 85 300 

Gottweig 262 n. 3 

Grenoble 197 325 n. 5 

270 320, 325 f. 

Heidelberg 262 n. 3 

Karlsruhe, Augieusis Lvii 207 

Lxxxi 239 

Lxxxiii 33, 330 

cv 140 

cix 207 n. 5 

cxix 40ff., 133, 137, 201ff., 

245, 305, 338, etc. 



Karlsruhe, Augiensis cl 332 n. 3 

ccxxxiii ... 344 

cciiiv 341 n. 3 

Koln : see Cologne 

Laon 273 296 n, 1 

Le Mans 229 340 

London, British Museum 

Royal I B xn 262 

Harl. 659 61 

,1772 262, 296 n. 1 

3102 340 

Add. 10546 262 

11852 262 

21914 334 

24142 262 

Lucca plut. I 1 344 

Luxemburg 135 (29) 334 

Lyons 403 (329) + 1964 (1840) ... 159 f. 

Madrid, Bibl. Nac. (Toledo 2. 1) ... 262 

Archivo Historico Nac. i 334 

Bibl. Acad. Hist. 44 254 

Manchester, John Eylands Library 
lat. 15 301 

Metz 134 296 n. 1 

225 316 

Milan, Ambros. A 138 sup 340 

L99 sup 254 

210 sup 206 n. 2 

Milan, Bibl. Brera AE xiv 9 344 

Monte Cassino48 330 

150 52 

£f.209 340 

Munich, Staatsbibl., lat. 4577 323 

6210 334 

6214 334 

6229 262, 293 

6238 337 

6436 262 n. 3 

9545 33, 329, 337 

13038 268, 286 ff., 310 f. 

14500 61 

18530 33, 329 

Munich, Universitatsbibl., Cod. MS., 

in fol. 12 293 f. 

Naples, Bibl. Naz. VI B 3 340 

viB 11 340 

Orleans 88(85) 330 

221 (193) 278 

Oxford, Bodl. Library 

Gr. Misc. 13 208 n. 2 

251 217 n. 1 

Barlow 4 33, 334 

Junius 25 301 

Laud. Lat. 108 262 

Laud. misc. 130 206 

350 344 

Auct. T. II 24 296 n. 1 

Oxford, Ball. Coll. Library 

157 40ff.,137ff.,213ff., 

228, 242, 245, 265, etc. 

Oxford, Merton Coll. Library 

26 ". 223ff., 265 

Paris, B. N. 

Fr. 22364 311 

Gr. 107 171, 262 

Lat. 254 262 

321 169 

335 262f. 

653 37fE.,45ff., 51, 59f., 

63, 131, 178 n. 1,245 ff. 

1759 52 

1760 52n.5 

1761 52 

1762 52n.5 

1763 52n.5 

1764 265.n. 2 

1853 21 u. 5, 22, 271, 294 ff. 

2341 334 

2392 272, 330f. 

2393 330 

2394 330 

2394A 330 

2395 331 

2409 339 

2412 •.. 339f. 

2452 340 

2709 236 n. 2, 295 n. 1 

5253 331 n. 1 

9380 262, 272, 342 f. 

9451 249 n. 4 

9525 21 n. 5, 272ff. 

10837 276f. 

10878 320 n. 2, 331 f. 

115.53 262 

11929 344 

12045 334 

12125 296 n. 1 

12289 296 n. 1, 331 

12290 331 

12303 340 

12309 62 n. 2 

13339 52 

13409 339 

15180 342f. 

17290 340 

nouv. acq. lat. 1460 265 n. 2 

2171 163, 164 n. 2, 262 n. 3 

Paris, Bibl. de 1' Arsenal 8407 337 

Paris, Bibl. Ste-Genevieve 10 ... 342 f. 

18 343 

Petrograd, F. v. i No. 17 51 

Puy 1 342 f. 

Eome, Vat. 

lat. 133 - 206 

615 340 

4950 16 n. 1 

5763 254 

5775 331 

7223 167 

95.30 331 



Kome. Vat. 

lat. 9546 331 

fragments : see Index 1 

pal. 234 41 

574 300 

reg. 9 262 

98 331 

Eome, Biblioteca Capitolare (Basili- 

cana) C 102 340 

C 103 340 

Eome, Bibl. Vallicelliana A 8 340 

B 6 262 

E 5 267 n. 9 

Eome, Bibl. Vittorio Emmanuele 

(Sess. Lviii) 163, 262 n. 3 

St Chef (lost) 320ff. 

St Gall, Stiftsbibl. 48 337 

73 28ff., 33, 36f., 45f., 

50 f., 59 n. 1, 62 f., 

129 239 

158 296 n. 1 

222 341 n. 3 

240 341 n. 3 

333 340 

424 334 

435 334 

728 233 

St Mihiel 16 335 n. 3 

St Omer 257 334 

St Paul in Carinthia 25. 3. 19 

( = xxv a 1) 207 n. 5 

Salisbury Cathedral Library 5 . . . 283 iJ. 

Troyes 32 312 

432 51 

486 311ff. 

521 311 n. 2 

523 312 

669 316 

Venice, Bibl. Marciana 10 208 n. 2 

Vercelli Cathedral (treasury) 167 f. 

Bibl. Capitol. 49 (xxxix) ... 332 n. 3 
CLXxxiii 15 n. 2 

Verona LI (49) 84 u. 3 

Lxxxix 2.54 

Vienna 1163 323 

1247 27 f., 30, 243, 

328 f., 336, .339 

Wolfenbiittel 4097 (= Weissenb. 

13) 239f. 

474 ( = Weissenb. 64) .. 254, 262 n. 3 

Wiiizburg mp. th. f. 12 25, 30, 243, 

280, 326 ff. 

Ziirich, Kantonsbibl. xii 334 

XXXII 334 

Lxxii 337 


Allen, P. S. 274 

Anderson, W. Blair 323 n. 2, 325 n. 3 

tBannister, H. M. 16 n. 1, 228 n. 1, 

294 n. 3, 296, 300, 311 n. 1, 314 
Barbeau, L. 334 n. 3 
Bernoulli, C. C. 274 n. 3, 275 u. 2 
Brewer, H. 51 n. 2, 120, 176 
Bruyne, D. de 156 
Clark, A. C. 215 u. 5 
Davidson, W. L. 283 u. 5 
Denk, J. 158 

tDorez, L. 274, 303 n. 5, 311 u. 1, 331 u. 1 
Esposito, M. 341 n. 1 
Flamm, H. 229 
Eraser, J. 344 n. 2 
Gibbons, H. A. 325 n. 3 
Heer, J. M. 229 
tHilberg, I. 224 n. 1 
tHolder, A. 204, 229 n. 2, 294 n. 3, 301, 

325 n. 1, 332 n. 3 
tKyd, A. H. 334 n. 2 
Lehmann, P. 267 n. 7, 286, 294 n. 1, 

300, 301 

tLiebaert, P. 296 

Lindsay, W. M. 294 n. 3, 301 

Loofs, F. 256 n. 2 

Lowe, E. A. 215 n. 5, 294 n. 3, 299 n. 2, 

300, 301. 340 n. 3 
tMayor, J. E. B. 32 u. 1 
Morin, G. 62, 1-56 n. 2, 322 n. 2, 334 n. 3 
Naish, J. P. 104 n. 3 
Omont, H. 245 n. 2, 294 n. 4, 325 u. 1, 

334 n. 3 
Eiggenbach, E. 274 n. 3 
Eobinson, J. A. 41 n.4, 283 n.4, 318n.l, 321 
Both, C. 275 n. 2 
Shepherd, J. F. 283 n. 1 
Smith, Alfred J. 83 n. 1, 265 
Souter, John B. 325 n. 3 
Souter. W. Clark 73 n. 1 
Turner, C. H. 15, 41, 203 f., 294 i\ 3 
Turner, E. J. 140 
Weir, E. 254 
White, H. J. 303 u. 5 
Wilmart, A. 303 nn. 5, 6, 334 n. 1 
Wordsworth, Chr. 286 n. 1 

^ Published work is not included here. 



bCSTi 1 




3 5002 03247 5175 

BS 2649 . P44 1922 1 

Pelagius's expositions of 
thirteen epistles of St.